Imagen de Portada

Not Just the Destination, But the Journey: Reasoning Traces Causally Shape Generalization Behaviors

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Wen, Pengcheng, Zhu, Yanxu, Sun, Jiapeng, Zhu, Han, Zhou, Yujin, Chan, Chi-Min, Han, Sirui, Guo, Yike
Formato:	Preprint
Publicado:	2026
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

AM$^3$Safety: Towards Data Efficient Alignment of Multi-modal Multi-turn Safety for MLLMs
por: Zhu, Han, et al.
Publicado: (2026)

ThinkPatterns-21k: A Systematic Study on the Impact of Thinking Patterns in LLMs
por: Wen, Pengcheng, et al.
Publicado: (2025)

LRAS: Advanced Legal Reasoning with Agentic Search
por: Zhou, Yujin, et al.
Publicado: (2026)

SafeMT: Multi-turn Safety for Multimodal Language Models
por: Zhu, Han, et al.
Publicado: (2025)

HKGAI-V1: Towards Regional Sovereign Large Language Model for Hong Kong
por: Han, Sirui, et al.
Publicado: (2025)

Glance-or-Gaze: Incentivizing LMMs to Adaptively Focus Search via Reinforcement Learning
por: Bai, Hongbo, et al.
Publicado: (2026)

J1: Exploring Simple Test-Time Scaling for LLM-as-a-Judge
por: Chan, Chi-Min, et al.
Publicado: (2025)

What, Whether and How? Unveiling Process Reward Models for Thinking with Images Reasoning
por: Zhou, Yujin, et al.
Publicado: (2026)

DC-W2S: Dual-Consensus Weak-to-Strong Training for Reliable Process Reward Modeling in Biological Reasoning
por: Chan, Chi-Min, et al.
Publicado: (2026)

Beyond Surface Structure: A Causal Assessment of LLMs' Comprehension Ability
por: Han, Yujin, et al.
Publicado: (2024)

SafeLawBench: Towards Safe Alignment of Large Language Models
por: Cao, Chuxue, et al.
Publicado: (2025)

Causal Stories from Sensor Traces: Auditing Epistemic Overreach in LLM-Generated Personal Sensing Explanations
por: Zhu, Shanshan, et al.
Publicado: (2026)

Journey Before Destination: On the importance of Visual Faithfulness in Slow Thinking
por: Uppaal, Rheeya, et al.
Publicado: (2025)

Structure-Augmented Reasoning Generation
por: Parekh, Jash Rajesh, et al.
Publicado: (2025)

Measuring Hong Kong Massive Multi-Task Language Understanding
por: Cao, Chuxue, et al.
Publicado: (2025)

Self-Guided Defense: Adaptive Safety Alignment for Reasoning Models via Synthesized Guidelines
por: Wang, Yuhang, et al.
Publicado: (2025)

RQ-RAG: Learning to Refine Queries for Retrieval Augmented Generation
por: Chan, Chi-Min, et al.
Publicado: (2024)

Learning While Staying Curious: Entropy-Preserving Supervised Fine-Tuning via Adaptive Self-Distillation for Large Reasoning Models
por: Wang, Hao, et al.
Publicado: (2026)

IndexMem: Learned KV-Cache Eviction with Latent Memory for Long-Context LLM Inference
por: Yang, Xintong, et al.
Publicado: (2026)

Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization
por: Wang, Boshi, et al.
Publicado: (2024)

Can Post-Training Transform LLMs into Causal Reasoners?
por: Chen, Junqi, et al.
Publicado: (2026)

Don't Command, Cultivate: An Exploratory Study of System-2 Alignment
por: Wang, Yuhang, et al.
Publicado: (2024)

More Data or Better Data? A Critical Analysis of Data Selection and Synthesis for Mathematical Reasoning
por: Zhao, Yike, et al.
Publicado: (2025)

Reasoning Shapes Alignment: Investigating Cultural Alignment in Large Reasoning Models with Cultural Norms
por: Wang, Yuhang, et al.
Publicado: (2025)

Graceful Forgetting in Generative Language Models
por: Jiang, Chunyang, et al.
Publicado: (2025)

Towards Advanced Mathematical Reasoning for LLMs via First-Order Logic Theorem Proving
por: Cao, Chuxue, et al.
Publicado: (2025)

FinMME: Benchmark Dataset for Financial Multi-Modal Reasoning Evaluation
por: Luo, Junyu, et al.
Publicado: (2025)

Importance Weighting Can Help Large Language Models Self-Improve
por: Jiang, Chunyang, et al.
Publicado: (2024)

RAS: Retrieval-And-Structuring for Knowledge-Intensive LLM Generation
por: Jiang, Pengcheng, et al.
Publicado: (2025)

Benchmarking Multi-National Value Alignment for Large Language Models
por: Shi, Weijie, et al.
Publicado: (2025)

Robust Pronoun Fidelity with English LLMs: Are they Reasoning, Repeating, or Just Biased?
por: Gautam, Vagrant, et al.
Publicado: (2024)

Doing Good or Doing Right? Exploring the Weakness of Commonsense Causal Reasoning Models
por: Han, Mingyue, et al.
Publicado: (2021)

Generative RLHF-V: Learning Principles from Multi-modal Human Preference
por: Zhou, Jiayi, et al.
Publicado: (2025)

KG-FPQ: Evaluating Factuality Hallucination in LLMs with Knowledge Graph-based False Premise Questions
por: Zhu, Yanxu, et al.
Publicado: (2024)

NaturalThoughts: Selecting and Distilling Reasoning Traces for General Reasoning Tasks
por: Li, Yang, et al.
Publicado: (2025)

LegalReasoner: Step-wised Verification-Correction for Legal Judgment Reasoning
por: Shi, Weijie, et al.
Publicado: (2025)

Tracing Facts or just Copies? A critical investigation of the Competitions of Mechanisms in Large Language Models
por: Campregher, Dante, et al.
Publicado: (2025)

Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning
por: Zhu, Jiapeng, et al.
Publicado: (2026)

Eliciting Causal Abilities in Large Language Models for Reasoning Tasks
por: Wang, Yajing, et al.
Publicado: (2024)

Unveiling Causal Reasoning in Large Language Models: Reality or Mirage?
por: Chi, Haoang, et al.
Publicado: (2025)