ChemCoTBench-V2 is a new rule-verifiable benchmark with 5,620 samples across 18 tasks that evaluates LLM chemical reasoning traces using deterministic chemistry rules and reference traces rather than final answers alone.
Outcome accuracy is not enough: Aligning the reasoning process of reward models
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
years
2026 4verdicts
UNVERDICTED 4roles
method 1polarities
use method 1representative citing papers
Skill-RM unifies heterogeneous reward criteria by modeling reward computation as dynamic execution of a reusable Reward-Evaluation Skill within an agent framework.
RoRo uses alternating optimization of a Rubricor and Judge to create process rewards from outcome-cost-process preference pairs, then combines them with outcome rewards via GRPO to train stepwise model routers that outperform baselines on five reasoning benchmarks.
The paper introduces the Proxy Compression Hypothesis as a unifying framework explaining reward hacking in RLHF as an emergent result of compressing high-dimensional human objectives into proxy reward signals under optimization pressure.
citing papers explorer
-
From Answers to States: Verifiable Process-Level Evaluation of Chemical Reasoning in Large Language Models
ChemCoTBench-V2 is a new rule-verifiable benchmark with 5,620 samples across 18 tasks that evaluates LLM chemical reasoning traces using deterministic chemistry rules and reference traces rather than final answers alone.
-
Skill-RM: Unifying Heterogeneous Evaluation Criteria via Agent Skill
Skill-RM unifies heterogeneous reward criteria by modeling reward computation as dynamic execution of a reusable Reward-Evaluation Skill within an agent framework.
-
Rubric-Guided Process Reward for Stepwise Model Routing
RoRo uses alternating optimization of a Rubricor and Judge to create process rewards from outcome-cost-process preference pairs, then combines them with outcome rewards via GRPO to train stepwise model routers that outperform baselines on five reasoning benchmarks.
-
Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges
The paper introduces the Proxy Compression Hypothesis as a unifying framework explaining reward hacking in RLHF as an emergent result of compressing high-dimensional human objectives into proxy reward signals under optimization pressure.