News without the noise
New AI papers from arXiv, picked daily and explained with their claims, benchmark results and code.
2566 articles · page 50 of 54 · The Research front page
The authors train a privileged teacher using environment rewards, then jointly train a skill-free student through reinforcement learning and token-level on-policy distillation. Across Qwen2.5 models from 1.5B to 7B parameters, RetireOPD reports substantial gains on ALFWorld and WebShop over reinforcement learning alone, with every student eventually outperforming its teacher.
Fan et al. test a lightweight coding harness while independently varying three components: planning, action space, and context management. On SWE-Bench Verified and Terminal-Bench 2.1, they find that context management mainly prevents overflow under tight token budgets, while the value of planning and predefined tools depends strongly on the underlying model’s capabilities.
Suleymanzade et al. propose looped flows, a method that trains recurrent neural networks for many-step inference using local denoising losses. This overcomes the difficulty of training early updates to support later ones when backpropagation is truncated. The authors report state-of-the-art accuracy among looped models on six reasoning benchmarks, including 58.8% on ARC-AGI-1 and 12.2% on ARC-AGI-2.
The authors propose a guided super-resolution approach for digital surface models (DSMs) using denoising diffusion, improving coarse 5 m DSMs to 0.5 m resolution by leveraging high-resolution optical imagery. Experiments on several Central European cities show the method produces DSMs with crisper building outlines and more detailed roof structures than conventional interpolation or filtering.
Hu et al. test whether long-horizon robot manipulation requires memory inside the action policy itself. On LIBERO-Mem, their architecture places all interaction history in a multimodal agent and achieves 76.3% average completion with an episodically stateless action model, versus 14.8% for the strongest reported baseline.
The authors introduce WORLDVIEW, a benchmark of 8,960 prompts spanning 15 languages and 31 language–context pairings, to examine how text-to-image systems rewrite requests before generation. Across DALL-E-3, Imagen-4, and GPT-Image-1.5, revisions marked non-Western and non-Anglophone settings more heavily than the United States and repeatedly reduced them to narrow, stereotypical vocabularies.
Chen et al. construct a reverse, label-free posterior from an explicit likelihood and compare it with agents’ conventional forward predictions. Across five LLM backbones, combining the two reasoning paths performed better than forward-only aggregation, particularly when agents disagreed.
Li et al. designed AgentZip to reduce the memory consumed when agent tasks spawn many related sandbox sessions. Across LLM training and inference workloads, the system reduced sandbox-owned memory by up to 8.7-fold while limiting the slowdown from aggressive compression to 1.40-fold.
The authors introduce a benchmark whose questions require an average of 12.1 intermediate conclusions and include non-text evidence that affects the reasoning process. The strongest evaluated system reached 43.1% Overall Accuracy but only 34.3% Strict Accuracy, exposing failures that final-answer scoring can obscure.
Jha et al. trained dense Transformers and mixture-of-experts models under varying data-repetition rates, domain mixtures, expert counts, and expert granularities. They find that MoE models lose performance more rapidly than dense models as examples are reused, with susceptibility tracking total parameter count rather than the number of parameters activated per token. Strong masking-based regularization preserves an MoE advantage beyond 64 repetitions, but does not match training on unique data.
Yu, Fang, and Chen propose Agent-Integrated Software, a pattern combining a conventional application core, direct user interaction, and a built-in agent. Their transition-system framework connects user-facing task semantics to application execution, while interaction contracts and continuous assurance specify when delegated actions and outcomes remain valid.
Gao et al. developed VikingRAG, a retrieval-augmented generation system that navigates both the meaning and structure of documents while retrieving additional evidence only when needed. Across real-world datasets, the authors report token consumption of 11.6%–51.9% of state-of-the-art baselines, falling to 5.1%–32.5% when retrieval traces are reused and multi-round search is invoked selectively.
Wang et al. introduce SWRouter, a method for routing each turn of a multi-turn conversation to a suitable large language model. Across dialogue benchmarks, the authors report higher evaluation accuracy than both the best individual model and a strong conversation-context routing baseline.
Amirmoshiri, Sahneh and Jangjou built an agentic platform that transforms heterogeneous Chemistry, Manufacturing and Controls documents into a queryable, provenance-linked knowledge graph. On 505 questions drawn from 38 reports for one Sanofi small-molecule program, the lexical retrieval layer performed strongly on direct questions but weakened on comparative and corpus-wide queries.
Singh and Priyam introduce ChurnBench, an open-source benchmark that models changing enterprise data as a timeline rather than a fixed snapshot. Their experiments show that scheduled refresh policies bound staleness by time-to-live (TTL), making cache age alone a poor predictor of freshness errors.
The authors developed Ecdysis, a framework for improving the runtime code and instructions that govern an LLM agent’s execution. Rather than revising a harness after each failed task, it analyzes batches of failures for recurring patterns and uses multiple diagnostic roles to propose refinements. In the reported experiments, this reduced training time by up to 1.84× while increasing reasoning accuracy by 18.56% over existing harness-evolution methods.
Lu et al. frame agent-skill optimization as a budget-constrained search over a changing pool of candidates. Across six benchmarks and three target models, their method delivered the highest average performance among the tested approaches while requiring 55–58% less optimization cost than SkillOpt.
Ji, Hyun and Chen introduce ORCH, a framework that builds task-specific hierarchies for embodied AI teams using principles from human organization theory. Across 25 wildfire-response missions and eight underlying language models, the authors report substantial gains in mission outcomes and execution efficiency, including when organizations were generated automatically by language models.
This week’s clearest signal is that agent reliability is a systems problem: unstable evaluators, hidden interface differences, incomplete security context, and poorly managed memory can overwhelm model-level improvements. The strongest papers replace coarse outcomes with repeated measurements, step-level evidence, structured state, and controls that survive component boundaries.
Lum and Zhang implemented a shared-memory abstraction in AIOS where agents record structured memories but the system kernel controls how those memories are retrieved and inserted into prompts. Across 1,800 trials with three language models, their approach substantially outperformed unmanaged and standard retrieval-based memory while approaching full-context performance at lower computational cost.
Liu et al. replace sampled reinforcement-learning updates with exact optimization over the complete, enumerable set of genomic tool combinations. Across five frozen reasoners and three benchmarks, their method outperformed GRPO in all 15 experimental settings, by 6.75 percentage points on average.
Marron and Barr propose Agentic Just-In-Time Software Construction, in which an application includes an AI agent that observes runtime behavior and modifies software logic or interfaces in response. Their demonstration presents trace-driven human–AI co-construction as an alternative to delivering software solely as a fixed, prebuilt artifact.
Li and Li introduce RD-Forget, a training-free memory framework that separates long-term storage from the evidence used to answer each query. Across five memory and reasoning settings, the authors report that query conditioning and selective suppression of superseded facts improve answers while preserving access to historical information.
The authors built a digital-advertising simulator in which injected interventions provide known labels for otherwise ambiguous diagnostic problems. On 235 held-out episodes, reinforcement learning with these synthesized rewards improved a Qwen3.5-35B-A3B agent’s full top-ranked attribution accuracy to 0.757, above all evaluated prompted baselines.
Zhang et al. introduce ConvMem, a training-free framework that uses prompted language models as query-dependent summarization kernels over segments of long documents. On two synthetic long-context, multi-hop question-answering benchmarks, the authors report better performance than training-free baselines and stronger out-of-distribution behavior than reinforcement-learning-trained memory methods.
Chen and colleagues introduce a benchmark for GUI workflows that span multiple operating systems rather than remaining on one device. Their evaluation finds that leading open-source GUI agents struggle to carry intermediate results across platforms, preserve context, and complete long chains of dependent actions.
Chen et al. introduce an interface that lets vision-language models make fine-grained physical decisions using semantic action units, which deterministic interpreters translate into commands for particular robots. The authors report that this approach transfers across tasks, environments, and robot embodiments while outperforming representative agent-based and vision-language-action baselines.
Ma and Kordjamshidi introduce CLAMP, a decoding framework that converts visual scene evidence and a symbolic action model into constraints on an embodied planner. Across three benchmarks, the authors report better object grounding and safety, with remaining errors concentrated in perception and incorrectly specified constraints.
Yu, Jang and Yu study how the structure of reinforcement-learning rollouts affects exploration in mathematical reasoning models. They report that their difficulty-adaptive, sentence-level branching method improves pass@k—the probability of finding a correct answer within k attempts—and consequently benefits test-time scaling.
Wang and colleagues evaluated seven AI systems through both their APIs and consumer chat interfaces on nine benchmarks covering general capability, social bias, and sycophancy. API testing produced accuracy scores 3.4 percentage points higher on average, and exposed API settings could not reliably reproduce interface behavior.
Schall et al. introduce Q2D-Web, a benchmark designed to reflect the first retrieval stage of agentic retrieval-augmented generation systems. Across lexical, dense, and late-interaction retrievers, model rankings remain broadly stable under different relevance-labeling methods but vary substantially by topic, language, and query type.
Li et al. introduce SkillAdam, a method for iteratively improving the natural-language instructions and procedural knowledge used by frozen language-model agents. Across seven short- and long-horizon benchmarks, the authors report more stable optimization, higher task performance, and fewer, less costly revision rounds than prior self-evolution methods.
The authors replace conventional generate-then-filter data synthesis with a loop that generates, verifies, and refines examples at multiple stages. A 4B-parameter model trained on 11K ToolLoop examples reached 86.40% accuracy on the Berkeley Function Calling Leaderboard and generalized to ACEBench with substantially less training data than the cited baseline.
Yang et al. developed MeClear to identify retrieved memories that impair an LLM agent, including outdated, misleading, or mutually conflicting evidence. Across ten long-dialogue memory pools, the framework achieved 85.9% target recall and an 82.3% task recovery rate, improving recovery by 25.5 percentage points over a leave-one-out baseline.
Duesterwald et al. evaluate reliability by running the same AppWorld task five times, exposing a substantial difference between average success and repeatable success. Their self-evolving framework stores targeted guidelines about unstable trajectory steps, increasing five-for-five task success by 16 percentage points on repeated tasks and 13 points on similar tasks.
Ben Amor et al. built RefactorPlatform, an open-source harness for comparing repository-scale refactoring agents under controlled conditions. Experiments across four model families show that AST-aware retrieval—organizing code by its syntactic structure—substantially improves success, while their multi-agent configuration trails a lean retrieval-augmented single agent.
Geng, Wang and Yuan developed a compact-memory pipeline that merges related memories online and packs retrieved information into the prompt as grouped units. On two long-horizon memory benchmarks, the authors report better quality–token trade-offs than several streaming and retrieval baselines, with the strongest results in the 2,000–5,000-token range.
Wang and colleagues developed TROVE, an orchestration method that learns reusable skills and outcome-dependent transitions from evaluated workflow traces. Rather than committing to a fixed plan or replanning from scratch, TROVE validates the route after each top-level skill and selectively edits only the continuation invalidated by new evidence.
Zheng and Yang identify “security-context discontinuity”: failures caused when individually sound agent controls lose or reinterpret security information as actions move between components. They formalize an end-to-end contract system and report that its full configuration blocked harmful effects in 2,560 injected attack instances while completing 700 benign tasks and escalating 200 ambiguous cases.
Gao et al. formed independent teams of agents from the same base model, allowed them to accumulate private notes over ten episodes, and then exchanged agents occupying matching roles. On held-out tasks, swaps had little effect on task performance but raised the communication required per unit of progress by 16% to 63%.
The authors introduce RoboSPA, a diagnostic benchmark for testing spatial reasoning and long-horizon planning in language-conditioned robotic manipulation. Across 280 task variants and 527,000 trajectories, representative vision-language-action models struggled with complex spatial relationships, precise execution, and planning that requires memory.
Goyal and Ray tested whether four forms of agent memory remained usable after changing the model that writes or reads them. Across 48 synthetic histories, fixed-schema knowledge graphs transferred with almost no accuracy change, whereas compressed notes were strongly coupled to migration direction and partially updated retrieval indexes recovered less than half the benefit of full re-embedding.
Shi and colleagues built CUA-Universe, an environment-to-data pipeline that converts desktop software into reproducible settings where agents can act through both graphical interfaces and command lines. Training a 9B model on its verified trajectories substantially improved reported task success while reducing interaction steps and token use across three evaluations.
The authors introduce LatentStream, a progressive latent working-memory framework for multimodal LLMs processing streaming video. Instead of storing historical clips in an external memory bank and retrieving them as extra visual context, LatentStream converts retrieved evidence into compact, fixed-length latent memory tokens that evolve over time. Experiments show the method outperforms prior store-and-retrieve approaches on both online and offline video understanding benchmarks.
Wang et al. propose TAHI (test-time adaptation through human-agent interaction), which leverages a user's cross-session interaction history to update agent context and weights while maintaining an evolving rubric that captures personal evaluation criteria. Applied to writing and visual creation across 30 individuals, the adapted agents outperformed non-adapted baselines on individual tasks, and part of the personalization gains transferred to other users.
The authors systematically benchmark three approaches to identity preservation in generative image models: encoding identity in the input context, using trainable subject-specific parameters (like LoRA), or maintaining a persistent identity layer. Their results show that persistent identity layers consistently reduce identity degradation across iterative edits, small subject scales, and multi-subject compositions, while preserving image quality and instruction adherence.
The authors introduce a causal taxonomy to separate deceptive behavior from deceptive mechanisms in language models, testing distinctions across controlled guessing-game and stock-trading experiments. They find that deceptive outputs can arise without corresponding deceptive mechanisms, while also providing causal evidence that an agent's knowledge about a recipient's information state directly affects deceptive preference.
The authors operationalize the importance of a chain-of-thought reasoning step as its advantage—the change in expected reward from including that step—estimated via Monte Carlo rollouts. They find that LLM judges asked to identify high-importance steps outperform a prevalence baseline but fall well short of a noise ceiling, and that fine-tuning a step-level critic improves performance for incorrect responses but remains far from ceiling for correct ones.