Imagen de Portada

Why Semantic Entropy Fails: Geometry-Aware and Calibrated Uncertainty for Policy Optimization

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Zhang, Zheyuan, Shi, Kaiwen, Bao, Han, Wang, Zehong, Ma, Tianyi, Ye, Yanfang
Formato:	Preprint
Publicado:	2026
Materias:	Machine Learning Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Can LLMs Convert Graphs to Text-Attributed Graphs?
por: Wang, Zehong, et al.
Publicado: (2024)

Policy4OOD: A Knowledge-Guided World Model for Policy Intervention Simulation against the Opioid Overdose Crisis
por: Ma, Yijun, et al.
Publicado: (2026)

Generative Graph Pattern Machine
por: Wang, Zehong, et al.
Publicado: (2025)

Why Reasoning Fails to Plan: A Planning-Centric Analysis of Long-Horizon Decision Making in LLM Agents
por: Wang, Zehong, et al.
Publicado: (2026)

LLM-Empowered Class Imbalanced Graph Prompt Learning for Online Drug Trafficking Detection
por: Ma, Tianyi, et al.
Publicado: (2025)

BHyGNN+: Unsupervised Representation Learning for Heterophilic Hypergraphs
por: Ma, Tianyi, et al.
Publicado: (2026)

Semantic Refinement with LLMs for Graph Representations
por: Thapaliya, Safal, et al.
Publicado: (2025)

Subgraph Pooling: Tackling Negative Transfer on Graphs
por: Wang, Zehong, et al.
Publicado: (2024)

Training MLPs on Graphs without Supervision
por: Wang, Zehong, et al.
Publicado: (2024)

SimMLP: Training MLPs on Graphs without Supervision
por: Wang, Zehong, et al.
Publicado: (2024)

OPBench: A Graph Benchmark to Combat the Opioid Crisis
por: Ma, Tianyi, et al.
Publicado: (2026)

Towards Graph Foundation Models: Learning Generalities Across Graphs via Task-Trees
por: Wang, Zehong, et al.
Publicado: (2024)

Beyond Message Passing: Neural Graph Pattern Machine
por: Wang, Zehong, et al.
Publicado: (2025)

AgentRouter: A Knowledge-Graph-Guided LLM Router for Collaborative Multi-Agent Question Answering
por: Zhang, Zheyuan, et al.
Publicado: (2025)

NGQA: A Nutritional Graph Question Answering Benchmark for Personalized Health-aware Nutritional Reasoning
por: Zhang, Zheyuan, et al.
Publicado: (2024)

GFT: Graph Foundation Model with Transferable Tree Vocabulary
por: Wang, Zehong, et al.
Publicado: (2024)

Harnessing Uncertainty: Entropy-Modulated Policy Gradients for Long-Horizon LLM Agents
por: Wang, Jiawei, et al.
Publicado: (2025)

Beyond Semantic Entropy: Boosting LLM Uncertainty Quantification with Pairwise Semantic Similarity
por: Nguyen, Dang, et al.
Publicado: (2025)

Drift-Bench: Diagnosing Cooperative Breakdowns in LLM Agents under Input Faults via Multi-Turn Interaction
por: Bao, Han, et al.
Publicado: (2026)

COPO: Consistency-Aware Policy Optimization
por: Han, Jinghang, et al.
Publicado: (2025)

Agentic Entropy-Balanced Policy Optimization
por: Dong, Guanting, et al.
Publicado: (2025)

EvolveRouter: Co-Evolving Routing and Prompt for Multi-Agent Question Answering
por: Huang, Jiatan, et al.
Publicado: (2026)

A Combinatorial Approach to Neural Emergent Communication
por: Zhang, Zheyuan
Publicado: (2024)

IAPO: Information-Aware Policy Optimization for Token-Efficient Reasoning
por: He, Yinhan, et al.
Publicado: (2026)

The Illusion of Certainty: Uncertainty Quantification for LLMs Fails under Ambiguity
por: Tomov, Tim, et al.
Publicado: (2025)

Latent Distribution Decoupling: A Probabilistic Framework for Uncertainty-Aware Multimodal Emotion Recognition
por: Huang, Jingwang, et al.
Publicado: (2025)

Interpretable Graph-Language Modeling for Detecting Youth Illicit Drug Use
por: Li, Yiyang, et al.
Publicado: (2025)

UNComp: Can Matrix Entropy Uncover Sparsity? -- A Compressor Design from an Uncertainty-Aware Perspective
por: Xiong, Jing, et al.
Publicado: (2024)

Asymmetric Advantage Modulation Calibrates Entropy Dynamics in RLVR
por: Gu, Hengrui, et al.
Publicado: (2026)

EPO: Entropy-regularized Policy Optimization for LLM Agents Reinforcement Learning
por: Xu, Wujiang, et al.
Publicado: (2025)

Temporal Graph Pattern Machine
por: Ma, Yijun, et al.
Publicado: (2026)

On the Safety of Graph Representation Learning
por: Guo, Xiaoguang, et al.
Publicado: (2026)

HawkesLLM: Semantic Uncertainty Propagation in Agentic Text Simulation
por: Deng, Zewei, et al.
Publicado: (2026)

Geometry-Calibrated Conformal Abstention for Language Models
por: Xu, Rui, et al.
Publicado: (2026)

Beyond the Score: Uncertainty-Calibrated LLMs for Automated Essay Assessment
por: Karim, Ahmed, et al.
Publicado: (2025)

Trained on Tokens, Calibrated on Concepts: The Emergence of Semantic Calibration in LLMs
por: Nakkiran, Preetum, et al.
Publicado: (2025)

CE-GPPO: Coordinating Entropy via Gradient-Preserving Clipping Policy Optimization in Reinforcement Learning
por: Su, Zhenpeng, et al.
Publicado: (2025)

Why Softmax Attention Outperforms Linear Attention
por: Deng, Yichuan, et al.
Publicado: (2023)

On the Entropy Calibration of Language Models
por: Cao, Steven, et al.
Publicado: (2025)

Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails
por: Frank, Gregory N.
Publicado: (2026)