group relative policy optimization

The papers

Most central to this idea first, not most recent.

• Independentcs.CL

  Dense rewards for language models lose to simpler reward when compute is matched

  Matched-compute comparisons show dense process rewards fail to beat baseline GRPO, while prompt selection yields a real gain.

  Nov 2026

• Top Universitycs.AI

  AI models can read pictured problems yet fail to solve them

  Multimodal LLMs fail to reason on images they can read because a reasoning-access gap blocks visual input from using text reasoning machinery.

  National University of Singapore, CNRS@CREATE · Nov 2026

• Top Universitycs.LG

  Merging specialist AI models saves 42% of training compute compared to joint training

  On-policy distillation beats GRPO in compute efficiency only when using the student's own GRPO checkpoint as teacher.

  Salesforce AI Research, UIUC · Nov 2026

• Big Techcs.LG

  Language models can match math accuracy with eight times fewer answers

  Latent-conditioned mutual information regularization resolves the exploration-convergence conflict in long-trace reasoning better than entropy bonuses.

  JD.com, Microsoft · Oct 2026

• Chinese Techcs.AI

  Fast generative models improve with RL when noise split is optimized geometrically

  Geo-FSRL selects stochastic policies minimizing conditional KL divergence, achieving higher reward scores than prior methods in few-step flow models.

  Beihang University, Sensetime Research · Nov 2026

• Chinese Techcs.CV

  AI video generation gains emotional fidelity when rewards trust only reliable cues

  EmoReward adaptively calibrates reward weights in emotional video generation using reliability learned from previous training run traces.

  The Hong Kong University of Science and Technology, Tencent Hunyuan · Nov 2026

• Top Universitycs.LG

  Retry-aware credit assignment boosts LLM reasoning efficiency and accuracy

  Retry-aware policy optimization reinforces successful retries without rewarding earlier failures, improving reasoning performance and token efficiency over GRPO.

  Columbia University, Yale University · Oct 2026

• Top Universitycs.CL

  LongZero trains LLMs on self-written long-context tasks, no external data

  LongZero improves LLM long-context reasoning via self-generated training tasks without any external data.

  Tsinghua University, Tsinghua-Bosch Joint ML Center · Oct 2026

• Chinese Techcs.AI

  Beacon restores safety signals lost during group-relative reasoning alignment

  GRPO safety alignment on reasoning models fails via advantage collapse and unsafe diversification; Beacon recovers safety signals, cutting safety error by up to 32.83 points.

  Baidu Inc., Zhongguancun Academy · Oct 2026

• Chinese Techcs.CL

  Learning the refinement loop improves web generation beyond final pages

  Learning generation, critique, and refinement jointly in a loop improves functional web generation more than optimizing each role in isolation.

  Tsinghua University, Huawei Noah's Ark Lab · Oct 2026

• Chinese Techcs.CV

  Varying sampled video frames strengthens reinforcement learning for temporal reasoning

  FrameGRPO improves video reasoning by varying frame sampling within rollout groups and rewarding successful sparse-view trajectories when full-frame rollouts fail.

  Shanghai Jiao Tong University, Shanghai Innovation Institute · Oct 2026

• Big Techcs.LG

  Token-level rewards make critic-free reasoning training faster and more accurate

  BoT-GRPO uses length-normalized token rewards to train reasoning models faster than GRPO, while improving React compilation, visual quality, and AIME results without a critic.

  Amazon AGI, Virginia Tech · Oct 2026

• Big Techcs.CV

  Separating visual and language reasoning phases improves vision-language model accuracy

  SPLIT-RL separates visual and language reasoning training stages with claim-level advantages, improving accuracy over GRPO by 1.4-6.1 points.

  University of Southern California, Amazon AGI · Oct 2026

• Big Techcs.LG

  Equal reward weighting can steer language models toward easier shortcuts

  Symmetric reward aggregation in LLM post-training degrades primary task performance by favoring easier auxiliary rewards.

  University of British Columbia, Amazon Annapurna Labs · Oct 2026

• Top Universitycs.CL

  RL gains in no-think mode stem from reactivating existing reasoning

  Reinforcement learning gains in no-think mode of hybrid LLMs largely come from reactivating the existing think mode reasoning.

  King Abdullah University of Science and Technology, Shanghai Jiao Tong University · Oct 2026

• Big Techcs.LG

  Training LLM forecasters across dates improves search and probability estimates

  Training an LLM forecaster to search evidence and report probabilities at multiple pre-resolution dates improves forecast accuracy and calibration.

  Amazon Data & Models, Georgia Institute of Technology · Oct 2026

• Big Techcs.AI

  Reinforcement learning adapts diffusion models to large routing problems

  ResizeCO adapts a diffusion model trained on small routing instances to large TSP and CVRP instances via reinforcement learning, setting a new state of the art for short run times.

  Bar-Ilan University, NVIDIA · Oct 2026

• Academiccs.CV

  Diagnose-to-Generate aligns flow models with localized structural corrections

  Diagnose-to-Generate (D2G) improves structural fidelity in flow-based image generators using patch-level feedback and trajectory-preserving stochastic sampling.

  University of Science and Technology of China, Meituan Inc. · Oct 2026

• Big Techcs.AI

  Agents search more effectively by choosing which reasoning branch to revisit

  TreeFlex, where one model picks which tree node to resume and what to do there, beats flat ReAct and FoldAgent on research and SWE tasks.

  Meta AI · Oct 2026

• Big Techcs.AI

  Turning failed reasoning attempts into training signal improves GRPO on hard problems

  Failed reasoning attempts can be turned into verified supervision for GRPO by having the policy write corrective notes and distilling only replays that beat the original group.

  University of California, San Diego, Adobe Research · Oct 2026

• Big Techcs.AI

  Learned memory assembly helps LLM agents succeed with fewer context tokens

  Learning to assemble memories with task success feedback improves LLM agent performance and reduces memory token usage.

  University of Illinois Urbana-Champaign, Amazon · Oct 2026

• Chinese Techcs.CV

  Continuous preference fields improve aesthetic crops beyond rigid annotation grids

  A continuous preference field for aesthetic cropping yields more accurate and generalizable crops than discrete grid annotation.

  Shanghai Jiao Tong University, Institute of Media Technology and Experience Design · Oct 2026

• Big Techcs.AI

  Simple credit shaping boosts exploration in language model reinforcement learning

  ExPPO redistributes reinforcement-learning credit toward surprising responses, improving reasoning coverage and correct-answer diversity while preserving verifier reward direction.

  Mila - Quebec AI Institute, Polytechnique Montréal · Oct 2026

• Industrycs.LG

  Difficulty-gated teacher guidance improves solution coverage in on-policy training

  DiffGate applies bounded teacher guidance to failed trajectories only, improving pass@8 over GRPO across Qwen3 students in code and math.

  Indian Institute of Science, Snap Inc. · Oct 2026

• Academiccs.CV

  Video RL improves by assigning credit to individual visual tokens

  TVRL uses gradients from a frozen vision-language model to allocate token-level credit, outperforming scalar-reward GRPO on video generation.

  Northeastern University · Oct 2026

• Big Techcs.CL

  Preference-grouped reinforcement learning improves social adaptation in simulated conversations

  PB-GRPO improves LLM social behavior by combining persona-driven simulation with preference-batched GRPO training.

  Amazon · Oct 2026

• Chinese Techcs.LG

  Sequence-level credit assignment improves on-policy distillation of reasoning models

  GR-OPD outperforms standard on-policy distillation by assigning sequence-level credit and calibrating with group-relative preferences.

  Shanghai Jiao Tong University, Alibaba Group · Oct 2026

• Big Techcs.AI

  Forecasting future actions makes language-agent training more sample-efficient

  Training agents to forecast their own future actions on successful trajectories improves agentic RL performance and sample efficiency without inference overhead.

  University of Illinois Urbana-Champaign, University of California San Diego · Oct 2026

• Chinese Techcs.AI

  KnowGPO aligns trajectories by knowledge for better credit assignment in search agents

  KnowGPO aligns trajectories by knowledge state, outperforming baselines by 3.8-4.2 points on deep search benchmarks.

  Zhejiang University, Tencent · Oct 2026

• Chinese Techcs.CL

  Specialized memory writer improves persistent state across multiple LLM frameworks

  A specialized memory writer trained with attributed chunk-level reinforcement learning outperforms general-purpose models across four long-term-memory harnesses.

  Tencent Hunyuan · Oct 2026

• Chinese Techcs.AI

  Adaptive curriculum training lets vision-language models solve hard visual Sudoku puzzles

  Adaptive curriculum learning via online rollout accuracy enables RLVR to solve long-horizon visual Sudoku puzzles where standard methods fail.

  Peking University, Huawei Technologies Co., Ltd. · Oct 2026

• Chinese Techcs.LG

  Calibrating group rewards can avoid failures from independent KL penalties

  ZCPO uses conditional-KL drift to calibrate zero-sum group-relative rewards, avoiding seven failure modes that can make independent reference-policy KL regularization harmful.

  Alibaba Group · Oct 2026

• Top Universitycs.LG

  Fisher-Rao prompt curricula make GRPO training more rollout-efficient

  ARCUS reduces GRPO rollouts by 48-57% and improves accuracy by 2.8-2.9 points using Fisher-Rao arc length for prompt selection.

  University of Wisconsin–Madison, University of Washington · Sept 2026

• Big Techcs.SD

  Chain-of-thought reasoning aligns speech translation duration with source audio

  DuraS2ST turns duration control in speech-to-speech translation into an explicit chain-of-thought planning problem solved with reinforcement learning.

  The Chinese University of Hong Kong, Microsoft Corporation · Oct 2026

• Research Labcs.LG

  HiLoRe allocates GRPO activations between storage, compression, and recomputation using update exposure.

  HiLoRe accelerates GRPO training by allocating activations among precise storage, low-precision compression, and recomputation according to their current update sensitivity.

  Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Tianjin Normal University · Sept 2026

• Chinese Techcs.CV

  Training world models for policy improvement beats optimizing pixel accuracy

  Training world models to maximize the policy's downstream return improves VLA policies even when next-frame pixel MSE is 17% worse.

  Hunyuan, Tencent, Peking University · Oct 2026

• Top Universitycs.CV

  Training models to recover regions from descriptions improves visual grounding

  A unified reinforcement learning loop couples region description and localization across images and videos, improving grounding without reference captions or external judges.

  NUS, UCAS · Oct 2026

• Chinese Techcs.CL

  Contrasting group executions pinpoints which agent skill caused success or failure

  GRSE compares multiple executions of the same question to assign credit to the responsible agent and skill, outperforming prior methods.

  Mayo Clinic, University of Manchester · Oct 2026

• Academiccs.LG

  Sparse probes improve credit assignment for language model reasoning

  ProbePO sparsely evaluates uncertain reasoning prefixes and redistributes GRPO credit, improving accuracy while cutting training time versus dense probing.

  Independent Researcher, University of Illinois Urbana-Champaign · Oct 2026

• Academiccs.CL

  Adaptive replay preserves reasoning diversity in RL fine-tuning

  Replay of diverse correct base-model solutions during RLVR preserves reasoning diversity without sacrificing performance.

  Cornell University · Oct 2026

• Chinese Techcs.LG

  Evidence annotations improve step-level credit for deep search agents

  Evidence-guided step-level process rewards derived from annotated evidence claims improve credit assignment and outperform standard GRPO on deep search benchmarks.

  Renmin University of China, Key Laboratory of Data Engineering and Knowledge Engineering, MOE · Oct 2026

• Chinese Techcs.CL

  Jointly learning what to store and retrieve improves agent memory

  Uni-Mem jointly learns memory construction and retrieval via end-to-end online RL, outperforming prior RL-based memory managers.

  Southeast University, Huawei Technologies, Noah’s Ark Lab · Oct 2026

• Industrycs.LGcode

  Training Forecasters to Search Improves Calibration and Reduces Tool Use

  Training language models to gather their own evidence during reinforcement learning improves forecasting and beats frontier models at lower cost.

  Future Principle · Oct 2026

• Chinese Techcs.LG

  Relative rubric comparisons improve rewards for open-ended language model training

  MatrixReward uses pairwise rubric comparisons to build a win-rate matrix and derive distance-based rewards for open-ended RL.

  Zhejiang University, Renmin University of China · Oct 2026

• Big Techcs.CV

  A unified evaluator scores images and pinpoints defects with explanations

  VIEScore2 jointly predicts image quality scores and spatially grounded defect locations, outperforming general-purpose VLMs on both tasks.

  University of Waterloo, NVIDIA · Oct 2026

• Big Techcs.LG

  Training on failed patches helps coding agents recover more effectively

  Training software engineering agents on recovery attempts conditioned on failed patches and verifier feedback improves resolution rates over standard GRPO.

  NVIDIA, Tsinghua University · Oct 2026

• Chinese Techcs.CV

  One OCR model family handles recognition, parsing, and reasoning

  PolyOCR achieves state-of-the-art performance across diverse OCR tasks by unifying instruction-following, a data engine, and competence-guided policy optimization.

  Ant Group · Oct 2026

• Chinese Techcs.CV

  Flow matching plus physics yields more trackable 3D human motion from video

  FlowHMR uses flow matching with reinforcement learning to recover physically plausible global 3D human motion from monocular video, beating prior methods in fidelity and tracking.

  Peking University, The Hong Kong University of Science and Technology · Oct 2026

• Big Techcs.AI

  GRPO Can Reward Lucky Guesses as If They Were Reasoning

  GRPO's advantage estimator rewards guessed correct answers, and SIGNBALANCE fixes this by decoupling magnitude from group statistics.

  Rochester Institute of Technology, Adobe Research · Sept 2026

• Chinese Techcs.LG

  Budgeting rewards per prompt helps language models train past reasoning plateaus

  Reward budgeting prevents premature convergence in reinforcement learning for LLM reasoning by assigning a fixed reward budget per prompt.

  University of Cambridge, Qwen Business Unit, Alibaba · Nov 2026

• Chinese Techcs.LG

  Separating search and answer credit improves accuracy and evidence grounding

  Separating search and answer credit improves both exact-match accuracy and evidence grounding in reinforcement-trained multi-turn search agents.

  South China University of Technology, FiT, Tencent · Oct 2026

• Top Universitycs.CL

  Mid-training stage internalizes safety without sacrificing general capability

  Safety mid-training reduces attack success rate by 72-81% while improving general capability.

  Johns Hopkins University, University of Washington · Oct 2026

• Academiccs.LG

  Pass@k misses output diversity changes caused by model post-training

  pass@k measures correct-answer probability but not output diversity, so it cannot detect post-training's effects on capability breadth.

  Accenture Strategy & Consulting, Vizuara AI Labs · Oct 2026

• Top Universitycs.LG

  Splitting an RLVR budget across multiple adapters outperforms a single adapter for majority voting

  Splitting an RLVR training budget across multiple LoRA adapters improves majority vote accuracy over training a single adapter.

  Princeton University · Oct 2026

• Chinese Techcs.LG

  Progress-aware advantage shaping resolves uniform credit assignment in agentic RL

  Progress-aware advantage shaping improves critic-free LLM-agent training by withholding rewards from invalid actions and concentrating penalties on deviations that cause downstream stagnation.

  Zhejiang University, Westlake University · Oct 2026

• Big Techcs.LGcode

  Formal process rewards improve language models on long agent tasks

  A Lean4-based framework that grounds multi-level process rewards in formal verification improves open-ended agentic RL for software engineering and terminal tasks.

  Amazon Web Services, University of Illinois Urbana-Champaign · Oct 2026

• Big Techcs.LG

  Reinforcement learning makes models’ admissions of failure hard to reproduce

  Failure disclosure in language models varies widely across reinforcement learning training runs, even when task accuracy remains stable.

  Stanford University, Anthropic · Sept 2026

• Chinese Techcs.LG

  Query-adaptive neuron steering improves frozen language models across benchmarks

  Q-NeuS adaptively modulates neurons per query to boost LLM performance without retraining or task labels.

  Northeastern University, Washington University in St. Louis · Oct 2026

• Industrycs.AI

  Computational graphs replace free-form reasoning for verifiable LLM outputs

  Training LLMs to generate executable computational graphs improves accuracy, cuts reasoning length 5-6x, and increases faithfulness versus chain-of-thought.

  MIPT, AIRI · Oct 2026

• Top Universitycs.LG

  Preserving multiple reasoning strategies prevents late-stage collapse in RLVR

  Reinforcement learning with verifiable rewards collapses because optimization concentrates probability onto one strategy, while accuracy requires multiple; Mesh Learning prevents this.

  Peking University · Oct 2026

• Chinese Techcs.CV

  MaskSense preserves subject exploration when training masked image generators with RL

  MaskSense uses semantic priors and entropy anchoring to avoid a visual exploration trap in reinforcement learning for masked image generation.

  University of Science and Technology of China, Tongyi Lab, Alibaba Group · Oct 2026

• Chinese Techcs.CL

  New RL method trains AI agents to handle multiple sessions effectively

  Mosaic uses hierarchical advantage optimization to outperform broadcasting in multi-session agentic RL on BrowseComp by 10.1% and with fewer sessions.

  HKUST (Guangzhou), Huawei · Oct 2026

• Chinese Techcs.CL

  Local retrieval credit helps search agents learn multi-step evidence gathering

  Local intermediate retrieval rewards improve search agent training more effectively than outcome-only reinforcement learning.

  University of Edinburgh, Huawei Technologies Research & Development (UK) Limited · Oct 2026

• Research Labcs.CL

  Accurate World Predictions Are Not Required for Agent Training Gains

  World model post-training improves agent performance through training effects other than accurate world prediction, as mismatched targets retain task gains despite lower accuracy.

  Xi’an Jiaotong University, University of Southern California · Sept 2026

• Academiccs.CL

  Training only the root token enables efficient adaptive reasoning

  Root-token policy optimization teaches reasoning models when to think while using about 2% of the training compute required by prior adaptive-reasoning methods.

  LG AI Research, UNIST · Oct 2026

• Chinese Techcs.CV

  Robotic world models learn by comparing diverse predicted futures

  FutureWorlds reduces LPIPS by 14.78%, 20.84%, and 9.12% on three datasets for 32-frame predictions.

  HKUST (GZ), CUHK · Oct 2026

• Chinese Tech

  Segment-specific rewards improve reinforcement learning for tool-calling agents

  Decoupling advantage estimation per segment resolves cross-segment credit misattribution in tool-calling RL.

  Peking University, Shenzhen University · Sept 2026

• Big Techcs.AI

  Value-guided search improves math reasoning during RL training

  APIVIS uses value-guided search during training to improve math reasoning beyond what independent sampling achieves.

  Amazon, Texas A&M University · Oct 2026

• Big Tech

  Privileged self-practice beats self-distillation for multi-turn language-model agents

  Privileged Self-Practice injects task-specific instructions into failing agent rollouts to improve multi-turn agent training.

  Texas A&M University, AWS AI, Amazon · Sept 2026

• Big Techcs.CL

  Teacher alignment method closes capacity gap in reasoning distillation without discarding data

  TeacherGRPO adapts teacher models to student capacity via reinforcement learning, closing the reasoning distillation gap curse.

  University of Virginia, Netflix · Oct 2026

• Big Techcs.CL

  Credit assignment that prunes redundant steps improves reasoning accuracy and efficiency

  RECAP assigns credit based on downstream dependency and correct-answer progress, improving accuracy and reducing reasoning tokens in large models.

  George Mason University, Amazon, Inc · Sept 2026

• Chinese Techcs.LG

  Reasoning models usually inherit “wait” from training data, not reinforcement learning

  The word "wait" in reasoning models is inherited from training data, not coined by reinforcement learning, which only reshapes it.

  The Hong Kong University of Science and Technology, Huawei Noahah’s Ark Lab · Oct 2026

• Research Labcs.LG

  Changing reward requirements teaches vision-language models to actually use images

  Multimodal reinforcement learning with verifiable rewards fails to learn visual skills unless tasks require visual evidence.

  University of the Chinese Academy of Sciences, Institute of Computing Technology, Chinese Academy of Sciences · Oct 2026

• Chinese Techcs.CV

  Spatial gradient maps improve RL alignment of flow-based image models

  SGA-Flow-GRPO uses spatial gradient-guided credit assignment to achieve state-of-the-art alignment quality for flow-based generative models on GenEval.

  Sun Yat-Sen University, Baidu · Sept 2026

• Top Universitycs.LG

  Drug discovery AI agents beat expert chemists at ranking molecule poses

  LLM-based agents using structured chemical features achieve up to 74% accuracy in ranking protein-ligand poses, exceeding expert chemists.

  Stanford University, Genesis Molecular AI · Nov 2026

• Chinese Techcs.CL

  Chatbots can learn to clarify vague requests without unnecessary questions

  DialogGRPO trains large language models to ask clarifying questions, recovering missing information on over 88% of vague queries.

  Zhejiang University, Ant Group · Nov 2026

• Chinese Techcs.CV

  Training with deceptive visual traps makes AI agents correct their own perception errors

  Training vision-language models with deliberately misleading visual descriptions, filtered so they change the answer, teaches them to self-correct errors and improves reasoning.

  Peking University, ByteDance · Nov 2026

• Chinese Techcs.AI

  Image generators follow object relationships more faithfully with targeted prompt edits

  StruPO uses calibrated scene graphs to learn prompt edits that improve image quality without changing the requested scene.

  Zhejiang University, Tencent AI Lab · Oct 2026

• Chinese Techcs.LG

  Language models solve more problems when training accounts for confidence

  SCARL reshapes advantages to align policy confidence with outcomes, improving reasoning accuracy and exploration in language model training.

  Baidu, Nanjing University · Oct 2026

• Big Techcs.CL

  AI jailbreak attacks more diverse with new training reward

  CoFiBreak trains jailbreak attackers to generate more varied successful attacks without sacrificing effectiveness or adding inference-time cost.

  Simon Fraser University, Microsoft Research · Nov 2026

• Chinese Techcs.MA

  Multi-agent AI systems improve by learning from execution traces

  MAS-Loop closes the feedback loop in multi-agent systems by using execution trajectories to refine and improve systems across training iterations.

  Hong Kong University of Science and Technology, Ant Group · Oct 2026

• Industrycs.RO

  Simulation training lets a pretrained model control robots on unfamiliar tasks

  A pretrained vision-language model can learn robot motion entirely in simulation and transfer to real tasks and a new robot embodiment.

  Luma AI · Oct 2026

• Chinese Techcs.LG

  AI agents can reuse successful decisions to learn with fewer tokens

  Replaying successful actions as branches inside rollout trees lets language-model agents learn long-horizon tasks with higher success and fewer rollout tokens.

  Alibaba Cloud Computing, Fudan University · Nov 2026

• Chinese Techcs.LG

  AI agents can recover forgotten skills from earlier training snapshots

  Agentic-OPD recovers lost capabilities from earlier RL checkpoints via on-policy distillation, improving agent performance beyond the best checkpoint.

  Baidu, Inc., Beijing Jiaotong University · Nov 2026

• Big Techcs.LG

  Language models can learn even when every sampled answer is wrong

  Steering vectors added to hidden states during training improve LLM post-training, especially when distilling from thinking teachers.

  Arizona State University, Amazon · Nov 2026

• Top Universitycs.CL

  Training AI from partially misaligned states makes behavior fixes last longer

  A new method, PEARL, keeps language models from relapsing into bad behaviors even after further training.

  Thinking Machines Lab, Princeton University · Nov 2026

• Big Techcs.CV

  Video AI answers 3D questions better when it predicts how objects move

  Teaching a video vision-language model to write out ordered 4D object states and reason over how they change between timesteps improves its answers to spatiotemporal questions.

  KAIST AI, Microsoft Research Asia · Nov 2026

• Top Universitycs.CL

  Teacher corrections with dense feedback double AI success on hard logic problems

  PIVOT nearly doubles mean avg@8 on hard logic games from 19.5% to 35.6%.

  University of Chinese Academy of Sciences, Tsinghua University · Oct 2026

• Chinese Techcs.LG

  Language models can learn from expert guidance when every attempt fails

  Policy-Compatible Reinforcement Learning recovers verified training examples for hard problems by guiding a model with expert solutions while limiting how far its behavior shifts.

  University of Science and Technology of China, Tencent · Oct 2026

• Chinese Techcs.CV

  Image generation improves through verified repairs without retraining the generator

  ImageRSI combines verification, experience library, and judge self-evolution to achieve 52.6% CSA on MindBench and 65.66% on ImageReq-Bench.

  University of Science and Technology of China, Qwen Business Unit of Alibaba, Alibaba Token Hub · Oct 2026

• Big Techcs.CL

  AI agents break chatbot safeguards more often in fewer exchanges

  An autonomous multi-turn jailbreak attacker can adapt to victim replies, decide when to stop, and achieve SOTA success rates without an in-loop judge.

  University of Rochester, Microsoft Research · Oct 2026

• Top Universitycs.AI

  Better static judges do not guarantee better training signals for generative reward models.

  Answer-only generative reward models outperform thinking models as online policy trainers despite lower static accuracy.

  SynthLabs.ai, Stanford University · Oct 2026

• Chinese Techcs.AI

  Targeting uncertain local decisions cuts GUI agent training cost by 56-61%

  A two-stage adaptive training framework for GUI agents reduces environment interaction by 56-61% while improving task performance.

  Alibaba Group, The Chinese University of Hong Kong · Oct 2026

• Big Techcs.LG

  LLM serving becomes batch-invariant with near-stock speed

  Detent achieves batch-invariant LLM serving and RL rollouts at near-stock speed by pinning only the batch-dependent values that affect kernel selection.

  Boston University, Google · Nov 2026

• Academiccs.LG

  Language models’ responses to reward changes can be forecast before training

  Within-prompt covariance between a response feature and a reward-model score change forecasts, before training, how far a policy's use of that feature will shift.

  Mohamed bin Zayed University of Artificial Intelligence · Nov 2026

• Top Universitycs.CV

  AI agents’ clicks marked wrong can work a third of the time

  Many clicks that GUI-grounding benchmarks reject are actually correct because they produce the same action as the annotated box.

  Zhejiang University, Zhejiang Geely Holding Group Co., Ltd · Oct 2026

• Research Labcs.AI

  Language models learn harder problems but can forget easier ones

  RLVR training primarily improves easy problem performance while expanding hard problem reasoning causes forgetting of easy skills.

  University of Notre Dame, The Pennsylvania State University · Nov 2026

• Chinese Techcs.CV

  AI models answer image questions better when taught where to look

  Rewarding zoom-in decisions directly, instead of only final answers, improves vision-language models' visual question answering.

  University of Science and Technology of China, Alibaba Group · Nov 2026

• Top Universitycs.LG

  Single-completion training matches multi-completion diversity by focusing on harder tasks

  Reweighting each prompt's regularization or reward turns ordinary single-sample reinforcement learning into an optimizer that shifts learning pressure toward harder problems and curbs sharpening.

  EPFL, ETH Zürich · Nov 2026

• Top Universitycs.LG

  Language models can improve coding answers without checking target correctness

  Test-Time Self-Play beats verifier-based methods for test-time scaling by training a solver on generated synthetic problems, improving by 35% with no verifier.

  ETH Zurich, Tufa Labs · Nov 2026

• Big Techcs.LG

  RL method helps AI agents learn high-level task strategies from sparse rewards

  A strategy-guided reinforcement learning method that routes LLM agents through hierarchical skills beats flat skill retrieval.

  Microsoft AI · Nov 2026

• Research Labcs.LG

  Coding agents solve more tasks when tool interfaces adapt during training

  Co-evolving an LLM agent's policy and its executable harnesses through reinforcement learning improves performance over optimizing either component alone.

  University of North Carolina at Chapel Hill, RIKEN · Oct 2026

• Research Labcs.CV

  Video generators can improve with feedback that lists visible flaws

  ArtiVidRM scores video by first listing its visible defects as a structured checklist, then using that defect list as a quality rubric.

  Southeast University, Opus AI Research · Nov 2026

• Chinese Techcs.AI

  AI agents complete more multistep tasks by reusing related experience

  Beyond-Local Policy Optimization improves long-horizon LLM agent training by reusing experience across tasks to estimate step-level credit without a critic.

  Beihang University, Alibaba Group · Nov 2026

• Industrycs.CL

  Shuffling reasoning steps yields better AI accuracy and faithfulness

  Scrambling reasoning steps and rewarding drops in answer probability improves model accuracy, reasoning quality, and faithfulness.

  UNC Chapel Hill · Nov 2026

• Top Universitycs.LGcode

  Language models find more valid solutions without supplied correct answers

  MS-TTRL resolves the frequency conflict by using self-verification for correctness and reserving generation frequency for diversity.

  Beihang University, National University of Singapore · Oct 2026

• Top Universitycs.CL

  Rewarding precision reduces unnecessary and incorrect content in rubric-trained models

  Adding a precision objective to rubric-based rewards reduces verbosity and improves factual correctness without sacrificing coverage.

  KAIST, Carnegie Mellon University · Oct 2026

• Academiccs.LG

  Self-Credit rescales self-assigned credit to improve long-horizon agentic RL

  Self-Credit improves long-horizon agentic RL by using model-cited critical turns rescaled across rollouts for dense credit assignment.

  Meta, Duke University · Oct 2026

• Chinese Techcs.LG

  Projecting conflicting gradients during joint RL and distillation improves agent training

  CORD improves joint RL and self-distillation by projecting conflicting gradients to preserve a protected objective.

  The Chinese University of Hong Kong, Tencent · Oct 2026

• Chinese Techcs.CV

  Verifiable rewards help video captioning models capture more complete visual narratives

  MVRCap frames video captioning as reinforcement learning with hierarchical multi-dimensional verifiable rewards, achieving state-of-the-art results on VDC, CAPability, and VC4VG.

  Xiamen University, Alibaba Group · Oct 2026

• Chinese Techcs.CL

  Adaptive rubrics improve reinforcement learning for open-ended text generation

  Selectively replacing saturated rubric criteria with flaw-focused checks improves reinforcement learning for open-ended generation while preserving useful feedback.

  State Key Laboratory of AI Safety, Institute of Computing Technology, CAS, University of Chinese Academy of Sciences · Oct 2026

• Big Techcs.CL

  Training search agents to assess missing evidence improves research performance

  Training retrieval-augmented LLMs with a dual-reward self-assessment phase improves search-state awareness and beats baselines on multi-hop and deep research benchmarks.

  Amazon, Korea University · Oct 2026

• Big Techcs.AI

  Consistency-regularized co-evolution of knowledge and model improves LLM agents

  LLM agents evolve more reliably by selecting experience-derived knowledge that yields consistent utility gains across multiple analysis views, rather than relying on any single interpretation.

  University of Illinois Urbana-Champaign, Georgia Institute of Technology · Oct 2026

• Top Universitycs.LG

  Small language models trained with reinforcement learning outperform 402B model in tool selection

  A Qwen3-8B model trained with reinforcement learning beats Llama-4-Maverick (402B) at tool selection and transfers to new tool corpora and retrievers.

  Meta, Carnegie Mellon University · Oct 2026

• Chinese Techcs.CV

  AI can trace objects’ outlines and flag when they’re absent

  A single vision-language model can handle detection, segmentation, matting, OCR, counting, and referring localization by predicting structured tokens plus a special object token that drives a pixel decoder.

  WeChat Vision, Tencent Inc. · Oct 2026

• Research Labcs.AI

  Learned latent links can undermine safety in multi-agent systems

  Training latent communication links can undermine otherwise safety-aligned agents, while reward-guided link optimization can both amplify and repair harmful compliance.

  CISPA Helmholtz Center for Information Security · Oct 2026

• Chinese Techcs.AI

  Solved mathematical mechanisms can generate verifiable training worlds for AI agents

  Generating agentic RL training environments from pre-solved mathematical mechanisms yields cheap, verifiable, stateful tasks that transfer to unseen and external benchmarks.

  Georgia Institute of Technology, Alibaba Token Foundry, Alibaba Group · Sept 2026

• Top Universitycs.LG

  Outcome-only reinforcement learning can teach models to judge reasoning steps

  Outcome-only RL can induce step-level process supervision in generative reward models, reaching 85.2 F1 on ProcessBench with 3.2K outcome-labeled trajectories.

  Renmin University of China, Tsinghua University · Oct 2026

• Big Techcs.CL

  Language models can cut reasoning tokens by targeting low-value text

  Normalized log probability identifies core and redundant tokens; compressing redundant tokens reduces token usage by 76% while preserving reasoning quality.

  Purdue University, Rochester Institute of Technology · Oct 2026

• Chinese Techcs.LG

  Training LLMs as adaptive solvers for industrial-scale optimization

  SDRL trains LLMs as adaptive meta-solvers that outperform DeepSeek-V4-Pro and GPT-5.5 on industrial-scale optimization.

  Tokentide AI, Alibaba Group · Oct 2026

• Big Techcs.CL

  Item response modeling improves rubric rewards while cutting judge requests

  Rubric Response Theory aggregates rubric criteria using item response theory to improve reinforcement learning rewards over additive methods.

  University of British Columbia, Microsoft · Sept 2026

• Chinese Techcs.LG

  Elastic GPU scheduling makes reinforcement learning faster for marathon agent tasks

  QwenGyre enables online reinforcement learning for xlong-horizon LLM agents by elastic GPU reallocation and trajectory deduplication.

  Alibaba Token Hub, Alibaba Group, University of Science and Technology of China · Sept 2026

• Big Techcs.AI

  Reinforcement learning teaches scientific ideation agents to explore broader possibilities

  AI Night-Scientist uses reinforcement learning to teach LLMs when to depart from predictable reasoning, improving scientific proposal diversity and originality.

  University of Illinois Urbana-Champaign, Microsoft · Sept 2026

• Top Universitycs.AIcode

  Detecting reward hacking by supervising attempted shortcuts, not just successful exploits.

  Monitoring internal states during code generation detects attempted reward hacking with high accuracy and low overhead, cutting cheating in RL from 82-91% to 1-5%.

  Nanyang Technological University, Independent Researcher · Oct 2026

• Chinese Techcs.AI

  Small test-time-trained guidance model beats adapting large LLM generators for discovery

  Guidance-TTT separates strategic guidance from solution execution, using a small test-time trained model to outperform prior methods.

  Massachusetts Institute of Technology, Hong Kong Polytechnic University · Oct 2026

• Chinese Techcs.CV

  Parallel tile inspection helps models find details in large images

  Visual Parallel Search with parallel tile inspection and adaptive zoom improves high-resolution VQA accuracy.

  The University of Hong Kong, Huawei Research · Oct 2026

• Top Universitycs.CL

  Attackers That Learn Mid-Attack Find More Large-Model Jailbreaks

  Updating the attacker's parameters during jailbreaking raises attack success rate from 55.9% to 72.4%.

  Johns Hopkins University, National University of Singapore · Oct 2026

• Chinese Techcs.AI

  LLM agents learn when to use atomic tools versus composite skills via counterfactual outcome comparisons

  CIPO improves LLM agents' tool use by training them to choose between atomic tools and composite skills based on state.

  Southeast University, KTH Royal Institute of Technology · Oct 2026

• Chinese Techcs.CL

  Rubric-based process rewards improve deep research agents’ evidence gathering

  Rubric-grounded process credit, which tracks accumulated support per rubric, improves deep research agent performance and enables an 8B model to match proprietary systems.

  University of Chinese Academy of Sciences, State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) · Sept 2026

• Top Universitycs.LG

  Shared rollout states improve credit assignment for long-horizon agents

  Cross-Rollout Bellman Closure improves long-horizon agentic RL by merging rollouts and computing a Bellman fixed point for step-level credit.

  Beihang University, Zhongguancun Academy · Sept 2026

• Research Labcs.LG

  Density-aware reweighting accelerates multi-reward RL for LLMs

  Inverse-square-root weighting by active-group density strengthens sparse reward signals and speeds multi-objective LLM reinforcement learning while preserving competitive final performance.

  University of Chinese Academy of Sciences, University of Minnesota Twin Cities · Oct 2026

• Chinese Techcs.CVcode

  Motion-space analysis exposes object movement flaws in generated videos

  MotionInsight diagnoses object motion deficiencies in generated videos by scoring object consistency, motion continuity, and physical plausibility from motion-space representations.

  MMLab, CUHK, ByteDance Inc. · Oct 2026

• Chinese Techcs.CV

  Video-grounded prompt planning improves long-form text-to-video generation quality

  WanPE enhances text-to-video generation by converting user prompts into cinematic plans, outperforming raw prompts and commercial systems.

  Nanjing University, Wan Team, Alibaba Group · Sept 2026

• Big Techcs.CL

  Verifying pivotal agent decisions improves reinforcement learning credit assignment

  Distributing credit only to pivotal decision segments improves large language model agent reinforcement learning outcomes.

  University of California, Davis, Microsoft · Sept 2026

• Chinese Techcs.CV

  Real-time GUI feedback improves training for computer-use agents

  ComputerSD uses real-time GUI feedback to improve computer-use agent training via online self-distillation.

  Zhejiang University, Ant Group · Oct 2026

• Big Techcs.AIcode

  Dynamic rubrics improve credit assignment for long-horizon agent training

  DRACO uses dynamic rubrics to assign per-step credit in GRPO, improving long-horizon agent training without ground-truth rewards.

  Carnegie Mellon University, IBM Research · Sept 2026

• Top Universitycs.AI

  Privileged supervision improves action-level credit for language-model agents

  TASPO converts privileged process supervision into outcome-grounded action credit, improving agentic policy optimization by 10.6% over GRPO across three benchmarks.

  Zhejiang University · Sept 2026

• Top Universitycs.AI

  Step-level checks curb unsafe agent actions with little utility loss

  StepGuard checks tool actions before execution and reduces attack success rate by 77.3% with only 2.8 percentage point utility drop.

  Shanghai Artificial Intelligence Laboratory, Beihang University · Aug 2026

• Chinese Techcs.CV

  Conversational image editing agent learns interpretable tool use

  Trained in three multitask alignment stages, IEA edits images conversationally by calling 16 interpretable tools step by step and outperforms baselines on fidelity, summary, and user preference.

  Shanghai Innovation Institution, Huawei Technologies Ltd. · June 2026

• Top Universitycs.AI

  Posterior rewards become indistinguishable for long reasoning traces in LLM training

  Probability-based reinforcement learning rewards for LLMs become indistinguishable for long reasoning traces, causing unstable training.

  National University of Singapore, The Hong Kong University of Science and Technology · Oct 2026

• Chinese Techcs.AI

  Human critiques help generative reward models produce more reliable evaluations

  EnGRICH uses human critiques to train a MetaCritic that improves generative reward model training with process supervision and guided exploration.

  Tsinghua University, Tencent · Oct 2026

• Big Techcs.LG

  Training the teacher on student prefixes boosts on-policy distillation performance

  Adapting the teacher to student-generated prefixes improves on-policy distillation.

  Washington University in St. Louis, AWS AI Labs · Oct 2026

• Chinese Techcs.LG

  Averaging rollout contexts improves label-free language model training

  Self-rewarding LLMs become more reliable when reward estimates are marginalized over random group contexts, improving zero-label self-evolution.

  Shanghai Jiao Tong University, Tencent · Sept 2026

• Big Techcs.IR

  Learned context curation boosts LLM catalog search accuracy

  Context curation and post-training with RPTune improves LLM catalog search accuracy by up to 31.4 percentage points.

  Google, University of Illinois Urbana-Champaign · Oct 2026

• Academiccs.LG

  Lean proof assistant generates step-level labels to train natural-language proof verifiers.

  Lean-generated step labels train better natural-language proof verifiers than human labels, without human annotation.

  University of Texas at Austin, University of California, Los Angeles · Oct 2026

• Top Universitycs.CV

  Training video models across counterfactual views improves evidence-grounded answers

  Optimizing video MLLMs on behavior packs across counterfactual views improves accuracy, temporal reasoning, and abstention over single-response post-training.

  Peking University, The University of Melbourne · Oct 2026

• Big Techcs.CV

  Co-evolving visual tasks keeps model training near its learning frontier

  Co-evolving visual environments with vision-language models via rollout pass rates improves out-of-domain reasoning by up to 5%.

  Virginia Tech, Cisco · Oct 2026

• Big Techcs.LG

  Context-aware credit assignment improves language agents on long-horizon tasks

  CCPO improves RL for LLM agents by weighting observation-matched peers by context and combining historical and future credit.

  Stanford University, Texas A&M University · Oct 2026

• Big Techcs.LG

  8B model learns to construct and refine GPU optimization scripts, beating larger LLMs

  A training method teaches an 8B language model to generate correct GPU optimization scripts, outperforming larger frontier LLMs.

  University of California, San Diego, Amazon Annapurna Labs · Oct 2026

• Chinese Techcs.LG

  Reinforcement-trained steering agent corrects goal drift in LLM agents and multi-agent systems

  A reinforcement-trained supervisor detects goal drift and selectively redirects LLM agents, improving single-agent and multi-agent performance with few interventions.

  National University of Singapore, University of Science and Technology of China · Oct 2026

• Chinese Techcs.CV

  Self-evolving AI agent generates cinematic multi-shot videos by iteratively directing and critiquing itself

  A unified Director Critic vision-language policy trained with alternating reinforcement learning improves cinematic multi-shot video generation.

  Beijing University of Posts and Telecommunications, Peking University · Oct 2026

• Big Techcs.AI

  Self-judged trajectories improve SWE agent training beyond binary verifier rewards

  Behavioral Self-Rewarding uses the policy's own trajectory comparisons to improve SWE agent training beyond binary test rewards.

  Pennsylvania State University, AG2AI · Oct 2026

• Chinese Techcs.LG

  Hindsight assigns token credit from one verified language-model rollout

  HP-RLVR assigns token credit from a single rollout per prompt, training 1.21x to 2.30x faster than GRPO with comparable accuracy.

  South China University of Technology, Tsinghua University · Oct 2026

• Big Techcs.AI

  Adaptive instruction selection helps language models learn from peers

  Adaptive instruction selection via a Stackelberg game improves multi-LLM collaborative alignment across benchmarks.

  University of Washington, Amazon · Oct 2026

• Big Techcs.LG

  Diverse supervised reasoning traces improve generalization after reinforcement learning

  Selecting diverse reasoning traces for SFT improves post-RL generalization on puzzles and mathematics.

  University of Illinois Urbana-Champaign, Google · Sept 2026

• Chinese Techcs.CLcode

  Tree-tracing reinforcement learning improves reasoning across long dialogue histories

  StateTree uses a tree-structured RL task to improve long-term dialogue reasoning, outperforming baselines on LongMemEval.

  Zhejiang University, Taobao & Tmall Group of Alibaba · Oct 2026

• Big Techcs.LG

  Explicit diversity training gives LLM agents distinct successful strategies

  Trajectory-guided Joint Policy Optimization (TJPO) improves task-relevant behavioral diversity in LLM agents.

  Microsoft, tuyoogame · Oct 2026

• Chinese Techcode

  Trajectory graphs improve step-level credit assignment in agentic reinforcement learning

  GRAFT uses trajectory graphs to estimate step-level advantages more faithfully than GRPO in multi-turn agentic tasks.

  Shanghai Jiao Tong University, Tencent AI Platform Department · Sept 2026

• Chinese Techcs.LG

  Online distillation cuts needless verbosity in RL-trained models

  Length Self-Distillation reduces excess response length on already-solved queries without sacrificing accuracy.

  ByteDance Seed, Tongji University · Oct 2026

• Chinese Techcs.CL

  Graded rewards improve language models’ control over scoped instructions

  ScopeIF uses graded rewards and a unified constraint schema to improve LLMs' precise instruction following on scope-aware constraints.

  Tsinghua University, Zhipu AI · Oct 2026

• Chinese Techcs.CV

  Interleaved table crops improve reasoning over dense visual layouts

  InterTab interleaves chain-of-thought with structure-aligned region crops to improve multimodal table reasoning.

  The Hong Kong University of Science and Technology (Guangzhou), Zhejiang University · Oct 2026

• Academiccs.LG

  Generating frontier-level problems improves reinforcement learning for LLM reasoning

  LLM reasoners improve by continually generating training problems at the edge of their evolving capability rather than using a fixed problem pool.

  University of Basel, University College London · Sept 2026

• Chinese Techcs.LG

  Historical trajectories provide step-level credit for training LLM agents via regressor

  T2SPO derives step-level credit from past successful trajectories using a frozen TabPFN regressor, improving LLM agent task success over GRPO.

  Nanjing University, ByteDance · Oct 2026

• Big Techcs.CLcode

  Paper-derived rubrics improve AI generation of scientific research plans

  Separating question sources from rubric criteria in paper-derived data cuts leakage to 3.7%; using the rubric as both teacher context and GRPO reward improves plan generation.

  Zhejiang University, Apple · Sept 2026

• Top Universitycs.AIcode

  Reinforcement learning strengthens policy invocation for agent safety judgments

  RePolicy learns to invoke safety policies for agent safeguards via reinforcement learning, achieving strong detection performance.

  Zhejiang University, Zhongguancun Academy · Aug 2026

• Top Universitycs.LG

  Language models can lag 50 to 150 updates behind training

  Standard bf16 rounding silently discards most RL updates, creating a hidden lag that unbiased rounding eliminates.

  Stanford University · Nov 2026

• Independentcs.RO

  World model simulator works across 22 robot platforms, enables policy learning

  PrimeSim turns a pretrained video generator into an action-conditioned, multiview world model that simulates robot manipulation across embodiments at 10.1 FPS.

  Nov 2026

• Big Techcs.CL

  Aligning auxiliary rewards with task success improves multi-task language model training

  CONCUR projects gated auxiliary rewards onto scores that agree with the primary reward, improving multi-task LLM reinforcement learning.

  CMU, Google DeepMind · Oct 2026

• Top Universitycs.LG

  Coding agents can learn neglected behaviors without sacrificing baseline performance

  Combining per-turn textual feedback self-distillation with RL stabilizes training and improves hard-to-optimize agent behaviors.

  Cursor Research, UC Berkeley · Nov 2026

• Top Universitycs.CL

  Language model RL failures can guide targeted pretraining repairs

  Failure-linked n-grams in RL rollouts can retrieve targeted pretraining data that improves reasoning more than equal-sized randomly sampled data.

  Princeton Language and Intelligence · Oct 2026

• Independentcs.CV

  Grounding foundation model surpasses larger vision-language models on precise perception and robotics

  GroundingPI, a 4B model, achieves state-of-the-art visual grounding across 34 benchmarks and improves robotic manipulation and autonomous driving when used as a perception backbone.

  Oct 2026

• Independentcs.CV

  Parallel diffusion decoding rivals autoregressive visual grounding at 4B scale

  GroundAnything uses blockwise diffusion to deliver faster parallel visual grounding while retaining strong localization accuracy across diverse tasks.

  Oct 2026

• Chinese Techcs.CV

  Multimodal models learn when to look and how long to reason

  L2T jointly optimizes the form and depth of computation for multimodal reasoning.

  Sun Yat-sen University, Shenzhen Loop Area Institute · Oct 2026

• Research Labcs.CL

  Generating user guides can improve web agents’ action selection

  Jointly decoding a guide and action improves web agent element selection, and guides causally drive actions.

  Independent Researcher, University of Chinese Academy of Sciences · Sept 2026

• Big Techcs.LG

  MEND selectively moves flow-model samples when reward gains justify distance

  MEND caps rewards per group and accepts a flow-model velocity update only when the capped reward gain exceeds a quadratic displacement price.

  The University of Texas at Austin, Google · Oct 2026

• Chinese Techcs.LG

  Verified intermediate progress improves reinforcement learning for tool-using agents

  ProCredit assigns credit to turns based on verified progress checks, improving agent training over outcome-only rewards.

  University of Chinese Academy of Sciences, Institute of Neuroscience, Chinese Academy of Sciences · Sept 2026

• Top Universitycs.CLcode

  Runtime research preserves details that prebuilt agent memories can lose

  Just-In-Time Agent Memory constructs query-relevant context at runtime, outperforming Ahead-of-Time systems and prior trained memory.

  Beijing Academy of Artificial Intelligence, Peking University · Sept 2026

• Chinese Techcs.LG

  Differentiable confidence readouts improve reasoning accuracy and calibration together

  CREDO improves reasoning accuracy and confidence calibration by replacing sampled confidence text with a deterministic, differentiable token-probability readout.

  University of Science and Technology of China, Qwen Business Unit of Alibaba · Sept 2026

• Top University

  LLM-discovered search operators adapt destruction and repair to search state

  SPO discovers adaptive destroy-repair programs for LNS using Stackelberg optimization and outperforms baselines on TSP and CVRP.

  CDL, Institute of Automation, Chinese Academy of Sciences, School of Artificial Intelligence, University of Chinese Academy of Sciences · Sept 2026

• Big Techcs.CL

  Distributed grounding preserves reasoning across million-token language model contexts

  DISCO separates parallel, sharded evidence grounding from centralized reasoning, preserving long-context accuracy while cutting inference cost versus monolithic full-context models.

  National University of Singapore, Adobe Research · Sept 2026

• Research Lab

  Reinforcement learning teaches an agent to verify mixed-source multimodal misinformation

  Reinforcement learning trains an LVLM to use a toolkit of specialized tools for detecting mixed-source multimodal misinformation, achieving 70.4% accuracy on MMFakeBench.

  Beijing University of Posts and Telecommunications, Minzu University of China · Sept 2026

• Chinese Techcs.CV

  Confidence scores for generated video intervals enable controllable selection without external verification

  Interval-level confidence scores trained with offline verifiers and reinforcement learning improve candidate selection in generative video temporal grounding.

  Alibaba Group, Beihang University · Oct 2026