Copertina

Patterns Over Principles: The Fragility of Inductive Reasoning in LLMs under Noisy Observations

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Li, Chunyang, Wang, Weiqi, Zheng, Tianshi, Song, Yangqiu
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Towards Multi-Agent Reasoning Systems for Collaborative Expertise Delegation: An Exploratory Design Study
di: Xu, Baixuan, et al.
Pubblicazione: (2025)

Rethinking Prospect Theory for LLMs: Revealing the Instability of Decision-Making under Epistemic Uncertainty
di: Wang, Rui, et al.
Pubblicazione: (2025)

Advancing Abductive Reasoning in Knowledge Graphs through Complex Logical Hypothesis Generation
di: Bai, Jiaxin, et al.
Pubblicazione: (2023)

The Cognitive Bandwidth Bottleneck: Shifting Long-Horizon Agent from Planning with Actions to Planning with Schemas
di: Xu, Baixuan, et al.
Pubblicazione: (2025)

Evaluating and Enhancing LLMs Agent based on Theory of Mind in Guandan: A Multi-Player Cooperative Game under Imperfect Information
di: Yim, Yauwai, et al.
Pubblicazione: (2024)

Can LLMs Learn to Reason Robustly under Noisy Supervision?
di: Yang, Shenzhi, et al.
Pubblicazione: (2026)

Controllable Logical Hypothesis Generation for Abductive Reasoning in Knowledge Graphs
di: Gao, Yisen, et al.
Pubblicazione: (2025)

Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning
di: Fan, Wei, et al.
Pubblicazione: (2026)

Inductive or Deductive? Rethinking the Fundamental Reasoning Abilities of LLMs
di: Cheng, Kewei, et al.
Pubblicazione: (2024)

LLM-Hanabi: Evaluating Multi-Agent Gameplays with Theory-of-Mind and Rationale Inference in Imperfect Information Collaboration Game
di: Liang, Fangzhou, et al.
Pubblicazione: (2025)

Acquiring and Modelling Abstract Commonsense Knowledge via Conceptualization
di: He, Mutian, et al.
Pubblicazione: (2022)

CodeGraph: Enhancing Graph Reasoning of LLMs with Code
di: Cai, Qiaolong, et al.
Pubblicazione: (2024)

WebDevJudge: Evaluating (M)LLMs as Critiques for Web Development Quality
di: Li, Chunyang, et al.
Pubblicazione: (2025)

DixitWorld: Evaluating Multimodal Abductive Reasoning in Vision-Language Models with Multi-Agent Dixit Gameplay
di: Mo, Yunxiang, et al.
Pubblicazione: (2025)

SciResearcher: Scaling Deep Research Agents for Frontier Scientific Reasoning
di: Zheng, Tianshi, et al.
Pubblicazione: (2026)

Enhancing Transformers for Generalizable First-Order Logical Entailment
di: Zheng, Tianshi, et al.
Pubblicazione: (2025)

Understanding Inter-Session Intentions via Complex Logical Reasoning
di: Bai, Jiaxin, et al.
Pubblicazione: (2023)

Agentic LLMs as Powerful Deanonymizers: Re-identification of Participants in the Anthropic Interviewer Dataset
di: Li, Tianshi
Pubblicazione: (2026)

LogiDynamics: Unraveling the Dynamics of Inductive, Abductive and Deductive Logical Inferences in LLM Reasoning
di: Zheng, Tianshi, et al.
Pubblicazione: (2025)

Mars: Situated Inductive Reasoning in an Open-World Environment
di: Tang, Xiaojuan, et al.
Pubblicazione: (2024)

Rethinking the Bounds of LLM Reasoning: Are Multi-Agent Discussions the Key?
di: Wang, Qineng, et al.
Pubblicazione: (2024)

Embeddings to Diagnosis: Latent Fragility under Agentic Perturbations in Clinical LLMs
di: Vijayaraj, Raj Krishnan
Pubblicazione: (2025)

Complex Reasoning over Logical Queries on Commonsense Knowledge Graphs
di: Fang, Tianqing, et al.
Pubblicazione: (2024)

Logical Reasoning with Relation Network for Inductive Knowledge Graph Completion
di: Zhang, Qinggang, et al.
Pubblicazione: (2024)

AutoGraph-R1: End-to-End Reinforcement Learning for Knowledge Graph Construction
di: Tsang, Hong Ting, et al.
Pubblicazione: (2025)

A Fragile Number Sense: Probing the Elemental Limits of Numerical Reasoning in LLMs
di: Rahman, Roussel, et al.
Pubblicazione: (2025)

Language Models as Inductive Reasoners
di: Yang, Zonglin, et al.
Pubblicazione: (2022)

Safety Compliance: Rethinking LLM Safety Reasoning through the Lens of Compliance
di: Hu, Wenbin, et al.
Pubblicazione: (2025)

KNOWCOMP POKEMON Team at DialAM-2024: A Two-Stage Pipeline for Detecting Relations in Dialogical Argument Mining
di: Zheng, Zihao, et al.
Pubblicazione: (2024)

On The Fragility of Benchmark Contamination Detection in Reasoning Models
di: Wang, Han, et al.
Pubblicazione: (2025)

NewtonBench: Benchmarking Generalizable Scientific Law Discovery in LLM Agents
di: Zheng, Tianshi, et al.
Pubblicazione: (2025)

Group Pattern Selection Optimization: Let LRMs Pick the Right Pattern for Reasoning
di: Wang, Hanbin, et al.
Pubblicazione: (2026)

INFERENCEDYNAMICS: Efficient Routing Across LLMs through Structured Capability and Knowledge Profiling
di: Shi, Haochen, et al.
Pubblicazione: (2025)

Message Intercommunication for Inductive Relation Reasoning
di: Liang, Ke, et al.
Pubblicazione: (2023)

Legal Rule Induction: Towards Generalizable Principle Discovery from Analogous Judicial Precedents
di: Fan, Wei, et al.
Pubblicazione: (2025)

Concept-Reversed Winograd Schema Challenge: Evaluating and Improving Robust Reasoning in Large Language Models via Abstraction
di: Han, Kaiqiao, et al.
Pubblicazione: (2024)

Inductive Reasoning for Temporal Knowledge Graphs with Emerging Entities
di: Zhao, Ze, et al.
Pubblicazione: (2026)

AbsPyramid: Benchmarking the Abstraction Ability of Language Models with a Unified Entailment Graph
di: Wang, Zhaowei, et al.
Pubblicazione: (2023)

Unifying Deductive and Abductive Reasoning in Knowledge Graphs with Masked Diffusion Model
di: Gao, Yisen, et al.
Pubblicazione: (2025)

GraphScout: Empowering Large Language Models with Intrinsic Exploration Ability for Agentic Graph Reasoning
di: Ying, Yuchen, et al.
Pubblicazione: (2026)