SOTAVerified

GSM8K

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 51–100 of 439 papers

TitleStatusHype
Synthetic Data RL: Task Definition Is All You NeedCode2
Data Whisperer: Efficient Data Selection for Task-Specific LLM Fine-Tuning via Few-Shot In-Context LearningCode1
SLOT: Sample-specific Language Model Optimization at Test-timeCode2
Reinforcing the Diffusion Chain of Lateral Thought with Diffusion Language Models—0
Accelerating Chain-of-Thought Reasoning: When Goal-Gradient Importance Meets Dynamic Skipping—0
AttentionInfluence: Adopting Attention Head Influence for Weak-to-Strong Pretraining Data Selection—0
S-GRPO: Early Exit via Reinforcement Learning in Reasoning Models—0
Elastic Weight Consolidation for Full-Parameter Continual Pre-Training of Gemma2—0
Rewriting Pre-Training Data Boosts LLM Performance in Math and CodeCode1
Memory-Efficient LLM Training by Various-Grained Low-Rank Projection of Gradients—0
Efficient Fine-Tuning of Quantized Models via Adaptive Rank and Bitwidth—0
TutorGym: A Testbed for Evaluating AI Agents as Tutors and StudentsCode0
NeMo-Inspector: A Visualization Tool for LLM Generation AnalysisCode1
Local Prompt Optimization—0
Trace-of-Thought Prompting: Investigating Prompt-Based Knowledge Distillation Through Question Decomposition—0
AutoJudge: Judge Decoding Without Manual Annotation—0
Efficient Reasoning for LLMs through Speculative Chain-of-ThoughtCode1
Training Large Language Models to Reason via EM Policy Gradient—0
Dynamic Early Exit in Reasoning ModelsCode2
Not All Rollouts are Useful: Down-Sampling Rollouts in LLM Reinforcement Learning—0
Entropy-Guided Watermarking for LLMs: A Test-Time Framework for Robust and Traceable Text Generation—0
Syzygy of Thoughts: Improving LLM CoT with the Minimal Free ResolutionCode3
Question Tokens Deserve More Attention: Enhancing Large Language Models without Training through Step-by-Step Reading and Question Attention Recalibration—0
Supervised Optimism Correction: Be Confident When LLMs Are Sure—0
Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use—0
SEAL: Steerable Reasoning Calibration of Large Language Models for FreeCode2
Sample, Don't Search: Rethinking Test-Time Alignment for Language Models—0
Sustainable LLM Inference for Edge AI: Evaluating Quantized LLMs for Energy Efficiency, Output Accuracy, and Inference Latency—0
Reasoning Under 1 Billion: Memory-Augmented Reinforcement Learning for Large Language ModelsCode0
Large (Vision) Language Models are Unsupervised In-Context LearnersCode1
Exploring LLM Reasoning Through Controlled Prompt VariationsCode0
Adaptive Rectification Sampling for Test-Time Compute ScalingCode0
Entropy-Based Adaptive Weighting for Self-TrainingCode1
CPPO: Accelerating the Training of Group Relative Policy Optimization-Based Reasoning ModelsCode2
Qwen2.5-Omni Technical ReportCode7
Lost in Cultural Translation: Do LLMs Struggle with Math Across Cultural Contexts?Code0
D^2LoRA: Data-Driven LoRA Initialization for Low Resource Tasks—0
Chain-of-Tools: Utilizing Massive Unseen Tools in the CoT Reasoning of Frozen Language ModelsCode2
SafeMERGE: Preserving Safety Alignment in Fine-Tuned Large Language Models via Selective Layer-Wise Model MergingCode1
Tapered Off-Policy REINFORCE: Stable and efficient reinforcement learning for LLMs—0
Improving Complex Reasoning with Dynamic Prompt Corruption: A soft prompt Optimization Approach—0
Towards Hierarchical Multi-Step Reward Models for Enhanced Reasoning in Large Language ModelsCode1
Rule-Guided Feedback: Enhancing Reasoning by Enforcing Rule Adherence in Large Language Models—0
Position-Aware Depth Decay Decoding (D^3): Boosting Large Language Model Inference Efficiency—0
SOLAR: Scalable Optimization of Large-scale Architecture for Reasoning—0
Self-Evolved Preference Optimization for Enhancing Mathematical Reasoning in Small Language Models—0
PromptCoT: Synthesizing Olympiad-level Problems for Mathematical Reasoning in Large Language ModelsCode1
DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning AbilityCode0
CODI: Compressing Chain-of-Thought into Continuous Space via Self-DistillationCode0
Layer-Aware Task Arithmetic: Disentangling Task-Specific and Instruction-Following Knowledge—0
Show:102550
← PrevPage 2 of 9Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1XolverAccuracy98.1—Unverified
2Orange-mini0-shot MRR98—Unverified