No article text was recorded for this version.
reasoning-outcome misalignment
Version 1 of 1223 papers275 conceptsfrozen 2026-10-04
Papers in this version
- Blog: The Unlearnability Phenomenon in RLVR for Language Models
- Blog: SophiaVL-R1: Reinforcing MLLMs Reasoning with Thinking Reward
- Blog: Is PRM Necessary? Problem-Solving RL Implicitly Induces PRM Capability in LLMs
- Blog: SAIL-RL: Guiding MLLMs in When and How to Think via Dual-Reward RL Tuning
- Blog: Do LLM Agents Know How to Ground, Recover, and Assess? Evaluating Epistemic Competence in Information-Seeking Agents
- Blog: Reasoning Models Hallucinate More: Factuality-Aware Reinforcement Learning for Large Reasoning Models
- Blog: Med-RLVR: Emerging Medical Reasoning from a 3B base model via reinforcement Learning
- Blog: Posterior-GRPO: Rewarding Reasoning Processes in Code Generation
- Blog: RFEval: Benchmarking Reasoning Faithfulness under Counterfactual Reasoning Intervention in Large Reasoning Models
- Blog: ReFT: Reasoning with Reinforced Fine-Tuning
- Blog: Enhancing the Outcome Reward-based RL Training of MLLMs with Self-Consistency Sampling
- Blog: Path Spuriousness-aware Reinforcement Learning for Multi-Hop Knowledge Graph Reasoning
- Blog: Limits of Generalization in RLVR: Two Case Studies in Mathematical Reasoning
- Blog: Step-Aware Policy Optimization for Reasoning in Diffusion Large Language Models
- Blog: PRMBench: A Fine-grained and Challenging Benchmark for Process-Level Reward Models
- Blog: Measuring Weak-to-Strong Legibility of Reasoning Models
- Blog: GRPO-CARE: Consistency-Aware Reinforcement Learning for Multimodal Reasoning
- Blog: Reasoning or Retrieval? A Study of Answer Attribution on Large Reasoning Models
- Blog: What If We Allocate Test-Time Compute Adaptively?
- Blog: Beyond Accuracy: Dissecting Mathematical Reasoning for LLMs Under Reinforcement Learning
- Blog: Reasoning Models Sometimes Output Illegible Chains of Thought
- Blog: Curing Miracle Steps in LLM Mathematical Reasoning with Rubric Rewards
- Blog: ChessArena: A Chess Testbed for Evaluating Strategic Reasoning Capabilities of Large Language Models
- Blog: Local Coherence or Global Validity? Investigating RLVR Traces in Math Domains
- Blog: SCE: Semantic Consistency Enhanced Reinforcement Learning for Multi-Hop Knowledge Graph Reasoning
- Blog: Towards Generalizable Reasoning: Group Causal Counterfactual Policy Optimization for LLM Reasoning
- Blog: Hacking Hallucinations of MLLMs with Causal Sufficiency and Necessity
- Blog: Nudging the Boundaries of LLM Reasoning
- Blog: Output Supervision Can Obfuscate the Chain of Thought
- Blog: Distributional Biases in Post-Training: A Markovian Analysis of Reasoning Trajectories
- Blog: RLVR Training of LLMs Does Not Improve Thinking Ability for General QA: Evaluation Method and a Simple Solution
- Blog: Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens
- Blog: StepORLM: A Self-Evolving Framework With Generative Process Supervision For Operations Research Language Models
- Blog: Reward Modeling from Natural Language Human Feedback
- Blog: Reject Only Critical Tokens: Pivot-Aware Speculative Decoding
- Blog: Measuring the Faithfulness of Thinking Drafts in Large Reasoning Models
- Blog: From System 1 to System 2: A Survey of Reasoning Large Language Models
- Blog: 7B Fully Open Source Moxin-LLM/VLM -- From Pretraining to GRPO-based Reinforcement Learning Enhancement
- Blog: Beyond Correctness: Harmonizing Process and Outcome Rewards through RL Training
- Blog: Beneficial Reasoning Behaviors in Agentic Search and Effective Post-training to Obtain Them
- Blog: Grounded Reinforcement Learning for Visual Reasoning
- Blog: PRM-PBE: Process Reward Model for Reinforcement Learning in Programming-by-Example
- Blog: Right Is Not Enough: The Pitfalls of Outcome Supervision in Training LLMs for Math Reasoning
- Blog: TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning
- Blog: To Think or Not To Think, That is The Question for Large Reasoning Models in Theory of Mind Tasks
- Blog: FAPO: Flawed-Aware Policy Optimization for Efficient and Reliable Reasoning
- Blog: Solver-Informed RL: Grounding Large Language Models for Authentic Optimization Modeling
- Blog: Learning Self-Correction in Vision–Language Models via Rollout Augmentation
- Blog: LLM Post-Training: A Deep Dive into Reasoning Large Language Models
- Blog: KnowRL: Exploring Knowledgeable Reinforcement Learning for Factuality
- Blog: Is PRM Necessary? Problem-Solving RL Implicitly Induces PRM Capability in LLMs
- Blog: Achieving Gold-Medal-Level Olympiad Reasoning via Simple and Unified Scaling
- Blog: Output Supervision Can Obfuscate the Chain of Thought
- Blog: Bridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning Alignment
- Blog: Thinking Forward and Backward: Multi-Objective Reinforcement Learning for Retrieval-Augmented Reasoning
- Blog: ExGRPO: Learning to Reason from Experience
- Blog: Reason-KE++: Aligning the Process, Not Just the Outcome, for Faithful LLM Knowledge Editing
- Blog: BOLT: Bootstrap Long Chain-of-Thought in Language Models without Distillation
- Blog: Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs
- Blog: GALAX: Graph-Augmented Language Model for Explainable Reinforcement-Guided Subgraph Reasoning in Precision Medicine
Showing 60 of 223 papers.
History & citation
Current · Permanent link
- This direction split from Process Supervision for Faithful Chain of Thought Reasoning (51% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_reasoning_outcome_misalignment_82f667b7,
title = {reasoning-outcome misalignment},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/reasoning-outcome-misalignment/rdr_82f667b76f7c4cc782ab80dcc0e5420b},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}This exact version (version 1, frozen text):
@misc{lacuna_reasoning_outcome_misalignment_82f667b7_v1,
title = {reasoning-outcome misalignment},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/reasoning-outcome-misalignment/rdr_82f667b76f7c4cc782ab80dcc0e5420b/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}