Cover Image

Revisiting LLM Reasoning via Information Bottleneck

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Lei, Shiye, Cheng, Zhihao, Jia, Kai, Tao, Dacheng
Format:	Preprint
Published:	2025
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Offline Behavioral Data Selection
by: Lei, Shiye, et al.
Published: (2025)

A Step Back: Prefix Importance Ratio Stabilizes Policy Optimization
by: Lei, Shiye, et al.
Published: (2026)

Offline Behavior Distillation
by: Lei, Shiye, et al.
Published: (2024)

State Diversity Matters in Offline Behavior Distillation
by: Lei, Shiye, et al.
Published: (2025)

EAPO: Enhancing Policy Optimization with On-Demand Expert Assistance
by: Song, Siyao, et al.
Published: (2025)

Image Captions are Natural Prompts for Text-to-Image Models
by: Lei, Shiye, et al.
Published: (2023)

MedCoG: Maximizing LLM Inference Density in Medical Reasoning via Meta-Cognitive Regulation
by: Zhao, Yu, et al.
Published: (2026)

Graph-Augmented Reasoning: Evolving Step-by-Step Knowledge Graph Retrieval for LLM Reasoning
by: Wu, Wenjie, et al.
Published: (2025)

Odysseus: Jailbreaking Commercial Multimodal LLM-integrated Systems via Dual Steganography
by: Li, Songze, et al.
Published: (2025)

Compose and Fuse: Revisiting the Foundational Bottlenecks in Multimodal Reasoning
by: Wang, Yucheng, et al.
Published: (2025)

LUMINA: LLM-Guided GPU Architecture Exploration via Bottleneck Analysis
by: Zhang, Tao, et al.
Published: (2026)

Robust Multimodal Sentiment Analysis via Double Information Bottleneck
by: Huang, Huiting, et al.
Published: (2025)

Revisiting the Uniform Information Density Hypothesis in LLM Reasoning
by: Gwak, Minju, et al.
Published: (2025)

Revisiting Catastrophic Forgetting in Large Language Model Tuning
by: Li, Hongyu, et al.
Published: (2024)

SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning
by: Pan, Rui, et al.
Published: (2025)

Revisiting Counterfactual Regression through the Lens of Gromov-Wasserstein Information Bottleneck
by: Yang, Hao, et al.
Published: (2024)

LLM Data Selection and Utilization via Dynamic Bi-level Optimization
by: Yu, Yang, et al.
Published: (2025)

MemFly: On-the-Fly Memory Optimization via Information Bottleneck
by: Zhang, Zhenyuan, et al.
Published: (2026)

Dynamic Graph Information Bottleneck
by: Yuan, Haonan, et al.
Published: (2024)

ReportBench: Evaluating Deep Research Agents via Academic Survey Tasks
by: Li, Minghao, et al.
Published: (2025)

Phenotypic Profile-Informed Generation of Drug-Like Molecules via Dual-Channel Variational Autoencoders
by: Liu, Hui, et al.
Published: (2025)

Mitigating Translationese Bias in Multilingual LLM-as-a-Judge via Disentangled Information Bottleneck
by: Zhang, Hongbin, et al.
Published: (2026)

Large Language Models as an Indirect Reasoner: Contrapositive and Contradiction for Automated Reasoning
by: Zhang, Yanfang, et al.
Published: (2024)

Protecting Your LLMs with Information Bottleneck
by: Liu, Zichuan, et al.
Published: (2024)

Dynamic Parallel Tree Search for Efficient LLM Reasoning
by: Ding, Yifu, et al.
Published: (2025)

Breaking the Exploration Bottleneck: Rubric-Scaffolded Reinforcement Learning for General LLM Reasoning
by: Zhou, Yang, et al.
Published: (2025)

Safety Reasoning with Guidelines
by: Wang, Haoyu, et al.
Published: (2025)

OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging
by: Wei, Yongxian, et al.
Published: (2025)

Revisiting the UID Hypothesis in LLM Reasoning Traces
by: Gwak, Minju, et al.
Published: (2025)

Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning
by: Zhang, Kongcheng, et al.
Published: (2025)

IBNorm: Information-Bottleneck Inspired Normalization for Representation Learning
by: Zou, Xiandong, et al.
Published: (2025)

Distillation Traps and Guards: A Calibration Knob for LLM Distillability
by: Zhan, Weixiao, et al.
Published: (2026)

Leveraging Knowledge Graphs and LLM Reasoning to Identify Operational Bottlenecks for Warehouse Planning Assistance
by: Parekh, Rishi, et al.
Published: (2025)

LEAD: Breaking the No-Recovery Bottleneck in Long-Horizon Reasoning
by: Pushkin, Denys, et al.
Published: (2026)

Towards Reasonable Concept Bottleneck Models
by: Kalampalikis, Nektarios, et al.
Published: (2025)

Revealing Combinatorial Reasoning of GNNs via Graph Concept Bottleneck Layer
by: Niu, Yue, et al.
Published: (2026)

DDO: Dual-Decision Optimization for LLM-Based Medical Consultation via Multi-Agent Collaboration
by: Jia, Zhihao, et al.
Published: (2025)

Escaping the Context Bottleneck: Active Context Curation for LLM Agents via Reinforcement Learning
by: Li, Xiaozhe, et al.
Published: (2026)

Graph Information Bottleneck for Remote Sensing Segmentation
by: Shou, Yuntao, et al.
Published: (2023)

Revisiting Model Interpolation for Efficient Reasoning
by: Wu, Taiqiang, et al.
Published: (2025)