会员服务
Measuring Chain-of-Thought Monitorability Through Faithfulness and Verbosity
Arxiv
0+阅读 · 10月31日
Reasoning Models Sometimes Output Illegible Chains of Thought
Arxiv
0+阅读 · 10月31日
Thought Branches: Interpreting LLM Reasoning Requires Resampling
Thought Branches: Interpreting LLM Reasoning Requires Resampling
Arxiv
0+阅读 · 10月31日
VCORE: Variance-Controlled Optimization-based Reweighting for Chain-of-Thought Supervision
Arxiv
0+阅读 · 10月31日
LLMs Can Covertly Sandbag on Capability Evaluations Against Chain-of-Thought Monitoring
Arxiv
0+阅读 · 10月31日
Improving Human Verification of LLM Reasoning through Interactive Explanation Interfaces
Arxiv
0+阅读 · 10月30日
Chain-of-Thought Hijacking
Arxiv
0+阅读 · 10月30日
Unified Multimodal Chain-of-Thought Reward Model through Reinforcement Fine-Tuning
Arxiv
0+阅读 · 10月29日
Parrot: A Training Pipeline Enhances Both Program CoT and Natural Language CoT for Reasoning
Arxiv
0+阅读 · 10月29日
FutureSightDrive: Thinking Visually with Spatio-Temporal CoT for Autonomous Driving
Arxiv
0+阅读 · 10月29日
NoisyGRPO: Incentivizing Multimodal CoT Reasoning via Noise Injection and Bayesian Estimation
Arxiv
0+阅读 · 10月29日
Testing Cross-Lingual Text Comprehension In LLMs Using Next Sentence Prediction
Arxiv
0+阅读 · 10月29日
Scaling Latent Reasoning via Looped Language Models
Arxiv
0+阅读 · 10月29日
A Pragmatic Way to Measure Chain-of-Thought Monitorability
Arxiv
0+阅读 · 10月28日
Beyond MCQ: An Open-Ended Arabic Cultural QA Benchmark with Dialect Variants
Arxiv
0+阅读 · 10月28日
参考链接
微信扫码咨询专知VIP会员
Top