MATH-500
Emerging58papers using it
2025first seen
Dataset Card for MATH-500 This dataset contains a subset of 500 problems from the MATH benchmark that OpenAI created in their Let's Verify Step by Step paper. See their GitHub repo for the source file: https://github.com/openai/prm800k/tree/main?tab=readme-ov-file#math-splits
Papers using MATH-500 (58)
- Seek in the Dark: Reasoning via Test-Time Instance-Level Policy Gradient in Latent SpaceKimi k1.5: Scaling Reinforcement Learning with LLMsKnowledge Distillation from Large Reasoning Models to Compact Student Models: A Case Study on the John O Bryan Mathematics CompetitionCliff Tokens: Identifying Single-Token Failure Triggers in LLM Mathematical ReasoningSteering LLM Thinking with Budget GuidanceReinforcement Learning for Reasoning in Large Language Models with One
Training ExampleStep-KTO: Optimizing Mathematical Reasoning through Stepwise Binary
FeedbackVTC-R1: Vision-Text Compression for Efficient Long-Context ReasoningDecoding as Optimisation on the Probability Simplex: From Top-K to Top-P (Nucleus) to Best-of-K SamplersLocally Confident, Globally Stuck: The Quality-Exploration Dilemma in Diffusion Language ModelsS0 Tuning: Zero-Overhead Adaptation of Hybrid Recurrent-Attention ModelsKVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in Reasoning TasksMerlin's Whisper: Enabling Efficient Reasoning in Large Language Models via Black-box Persuasive PromptingBeyond Correctness: Learning Robust Reasoning via TransferCan I Have Your Order? Monte-Carlo Tree Search for Slot Filling Ordering in Diffusion Language ModelsDeCoVec: Building Decoding Space based Task Vector for Large Language Models via In-Context LearningTERMINATOR: Learning Optimal Exit Points for Early Stopping in Chain-of-Thought ReasoningTAPS: Task Aware Proposal Distributions for Speculative SamplingTowards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought CompressionAligning Tree-Search Policies with Fixed Token Budgets in Test-Time Scaling of LLMsLong Chain-of-Thought Compression via Fine-Grained Group Policy OptimizationReevaluating Self-Consistency Scaling in Multi-Agent SystemsPrompting Test-Time Scaling Is A Strong LLM Reasoning Data AugmentationFrom Implicit Exploration to Structured Reasoning: Leveraging Guideline and Refinement for LLMsMemLens: Uncovering Memorization in LLMs with Activation TrajectoriesMoL-RL: Distilling Multi-Step Environmental Feedback into LLMs for Feedback-Independent ReasoningFast on the Easy, Deep on the Hard: Efficient Reasoning via Powered Length PenaltyProcess Reward Models That ThinkThinkless: LLM Learns When to ThinkNot All Correct Answers Are Equal: Why Your Distillation Source MattersFractional Reasoning via Latent Steering Vectors Improves Inference Time
ComputeInpainting-Guided Policy Optimization for Diffusion Large Language
ModelsSocratic-Zero : Bootstrapping Reasoning via Data-Free Agent Co-evolutionPairwise RM: Perform Best-of-N Sampling with Knockout TournamentCan 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time
ScalingThinking Preference OptimizationSIFT: Grounding LLM Reasoning in Contexts via StickersThink Twice: Enhancing LLM Reasoning by Scaling Multi-round Test-time
ThinkingSample, Don't Search: Rethinking Test-Time Alignment for Language ModelsT1: Tool-integrated Self-verification for Test-time Compute Scaling in
Small Language ModelsSkywork R1V: Pioneering Multimodal Reasoning with Chain-of-ThoughtDianJin-R1: Evaluating and Enhancing Financial Reasoning in Large
Language ModelsPhi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language
Models in MathSeek in the Dark: Reasoning via Test-Time Instance-Level Policy Gradient
in Latent SpacePreMoe: Lightening MoEs on Constrained Memory by Expert Pruning and
RetrievalHarnessing Negative Signals: Reinforcement Distillation from Teacher
Data for LLM ReasoningConfidence Is All You Need: Few-Shot RL Fine-Tuning of Language ModelsReasonFlux-PRM: Trajectory-Aware PRMs for Long Chain-of-Thought
Reasoning in LLMsAgentar-Fin-R1: Enhancing Financial Intelligence through Domain
Expertise, Training Efficiency, and Advanced ReasoningScaleDiff: Scaling Difficult Problems for Advanced Mathematical
ReasoningQeRL: Beyond Efficiency -- Quantization-enhanced Reinforcement Learning
for LLMsReasoning with Sampling: Your Base Model is Smarter Than You ThinkWhen to Ensemble: Identifying Token-Level Points for Stable and Fast LLM
EnsemblingCan LLMs Guide Their Own Exploration? Gradient-Guided Reinforcement Learning for LLM ReasoningFaster and Better LLMs via Latency-Aware Test-Time ScalingWalk Before You Run! Concise LLM Reasoning via Reinforcement LearningAdaptive Rectification Sampling for Test-Time Compute ScalingControlling Large Language Model with Latent Actions