Search

6 results for “MATH-500”

1. Research · Language models

   Sparse RL-Guided Token Choices Recover Most Math Reasoning Gains

   What RL Brings to Reasoning: A Few Decisions the Base Model Would Not Make

   Replacing about 4% of a base model’s generated tokens with an RL-trained model’s choices recovered 90.5% of the reinforcement-learning improvement on AIME.

   AIME 2022-2025: 61.45% pass@256 (as reported)

   20 Oct 2026University of California, San Diego, Meituan, LongCat Team, The Chinese University of Hong Kong, Shenzhen

2. Research · Language models

   Cross-feedback and step-wise scoring let small models match much larger reasoning datasets

   Collaborative Reasoning Distillation via Cross-Feedback and Coherent Curation

   The authors' CRD framework iteratively critiques multiple teachers, scores individual reasoning steps, and stitches complementary chains to train compact models on 50K examples.

   MATH-500: 94.5% pass@1 (as reported)

   9 Oct 2026KAIST

3. Research · Language models

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

   Inherited, Not Invented: Tracing the Provenance of Wait in Reasoning Models

   Across open-model training histories, the authors find that reinforcement learning can improve mathematics accuracy without introducing “wait”, while different training recipes amplify or erase an inherited marker.

   MATH-500: 10.8% accuracy (as reported)

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

4. Research · Evaluation & benchmarks

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

   What pass@k Cannot Measure: Evaluating Diversity and Capability Retention after Post-Training

   GRPO and rejection-sampling fine-tuning produced sharply different output distributions despite often similar pass@k scores.

   GSM8K: 0.51 distinct-3 (verifier-correct completions) (as reported)

   8 Oct 2026Accenture Strategy & Consulting, Vizuara AI Labs

5. Research · Reinforcement learning

   Posterior rewards become indistinguishable for long reasoning traces in LLM training

   Rethinking Probability-Based Reinforcement Learning From Posterior Concentration

   The authors identify a length-dependent failure mode they call Posterior Concentration Phenomenon, and propose a framework to mitigate it.

   MATH-500: 82.9% Avg@2 (as reported)

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

6. Research · Reinforcement learning

   Preserving multiple reasoning strategies prevents late-stage collapse in RLVR

   All Work And No Play Makes Jack a Dull Boy: Understanding and Preventing Catastrophic Strategy Collapse in RLVR

   The authors connect abrupt performance loss to strategy concentration, derive an early-warning metric and test a strategy-balancing training method.

   MATH-500: 73.4% Mean@4 Accuracy (as reported)

   7 Oct 2026Peking University