Cover Image

Diagnosing Multi-step Reasoning Failures in Black-box LLMs via Stepwise Confidence Attribution

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Liu, Xiaoou, Chen, Tiejin, Zhang, Dengjia, Wang, Yaqing, Cheng, Lu, Wei, Hua
Format:	Preprint
Published:	2026
Subjects:	Computation and Language Artificial Intelligence Information Theory Machine Learning 68T50, 68T37, 68Q32 I.2.7; I.2.6; I.2.4
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Position: Uncertainty Quantification in LLMs is Just Unsupervised Clustering
by: Chen, Tiejin, et al.
Published: (2026)

Understanding the Uncertainty of LLM Explanations: A Perspective Based on Reasoning Topology
by: Da, Longchao, et al.
Published: (2025)

Rule Extraction in Machine Learning: Chat Incremental Pattern Constructor
by: Nwokocha, Caleb Princewill
Published: (2022)

Modularity in Transformers: Investigating Neuron Separability & Specialization
by: Pochinkov, Nicholas, et al.
Published: (2024)

CogniLoad: A Synthetic Natural Language Reasoning Benchmark With Tunable Length, Intrinsic Difficulty, and Distractor Density
by: Kaiser, Daniel, et al.
Published: (2025)

Superior Scoring Rules for Probabilistic Evaluation of Single-Label Multi-Class Classification Tasks
by: Ahmadian, Rouhollah, et al.
Published: (2024)

NOTAI.AI: Explainable Detection of Machine-Generated Text via Curvature and Feature Attribution
by: Breneur, Oleksandr Marchenko, et al.
Published: (2026)

Enhancing Ultra-Low-Bit Quantization of Large Language Models Through Saliency-Aware Partial Retraining
by: Cao, Deyu, et al.
Published: (2025)

JAM: Controllable and Responsible Text Generation via Causal Reasoning and Latent Vector Manipulation
by: Huang, Yingbing, et al.
Published: (2025)

On measuring grounding and generalizing grounding problems
by: Quigley, Daniel, et al.
Published: (2025)

CELI: Controller-Embedded Language Model Interactions
by: Wagner, Jan-Samuel, et al.
Published: (2024)

Rethinking the Multilingual Reasoning Gap with Layer Swap
by: Lasbordes, Maxence, et al.
Published: (2026)

D-SMART: Enhancing LLM Dialogue Consistency via Dynamic Structured Memory And Reasoning Tree
by: Lei, Xiang, et al.
Published: (2025)

How much do LLMs learn from negative examples?
by: Hamdan, Shadi, et al.
Published: (2025)

Communicative Agents for Slideshow Storytelling Video Generation based on LLMs
by: Fan, Jingxing, et al.
Published: (2025)

From Noise to Diversity: Random Embedding Injection in LLM Reasoning
by: Kim, Heejun, et al.
Published: (2026)

InhibiDistilbert: Knowledge Distillation for a ReLU and Addition-based Transformer
by: Zhang, Tony, et al.
Published: (2025)

CLMN: Concept based Language Models via Neural Symbolic Reasoning
by: Yang, Yibo
Published: (2025)

Think Thrice Before You Speak: Dual knowledge-enhanced Theory-of-Mind Reasoning for Persuasive Agents
by: Ma, Minghui, et al.
Published: (2026)

Conformal Path Reasoning: Trustworthy Knowledge Graph Question Answering via Path-Level Calibration
by: Lin, Shuhang, et al.
Published: (2026)

Mitigating LLM Hallucinations through Domain-Grounded Tiered Retrieval
by: Haque, Md. Asraful, et al.
Published: (2026)

NeuroState-Bench: A Human-Calibrated Benchmark for Commitment Integrity in LLM Agent Profiles
by: Jia, Xiao
Published: (2026)

Sliced-Wasserstein Distribution Alignment Loss Improves the Ultra-Low-Bit Quantization of Large Language Models
by: Cao, Deyu, et al.
Published: (2026)

Harnessing non-adversarial robustness in large language models
by: Zhou, Qinghua, et al.
Published: (2026)

When Does Content-Based Routing Work? Representation Requirements for Selective Attention in Hybrid Sequence Models
by: Basu, Abhinaba
Published: (2026)

Learning What Matters: Probabilistic Task Selection via Mutual Information for Model Finetuning
by: Chanda, Prateek, et al.
Published: (2025)

Constitution or Collapse? Exploring Constitutional AI with Llama 3-8B
by: Zhang, Xue
Published: (2025)

XAutoLM: Efficient Fine-Tuning of Language Models via Meta-Learning and AutoML
by: Estevanell-Valladares, Ernesto L., et al.
Published: (2025)

Thinking Longer, Not Always Smarter: Evaluating LLM Capabilities in Hierarchical Legal Reasoning
by: Zhang, Li, et al.
Published: (2025)

Deterministic Event-Graph Substrates as World Models for Counterfactual Reasoning
by: Rovai, Fabio
Published: (2026)

The Current and Future Perspectives of Zinc Oxide Nanoparticles in the Treatment of Diabetes Mellitus
by: Yousaf, Iqra
Published: (2024)

Vis-CoT: A Human-in-the-Loop Framework for Interactive Visualization and Intervention in LLM Chain-of-Thought Reasoning
by: Pather, Kaviraj, et al.
Published: (2025)

Mitigating Position-Shift Failures in Text-Based Modular Arithmetic via Position Curriculum and Template Diversity
by: Yudin, Nikolay
Published: (2026)

Natural Language Processing: A Comprehensive Practical Guide from Tokenisation to RLHF
by: Arabov, Mullosharaf K.
Published: (2026)

GE-Chat: A Graph Enhanced RAG Framework for Evidential Response Generation of LLMs
by: Da, Longchao, et al.
Published: (2025)

The Geometry of Persona: Disentangling Personality from Reasoning in Large Language Models
by: Wang, Zhixiang
Published: (2025)

Kodezi Chronos: A Debugging-First Language Model for Repository-Scale Code Understanding
by: Khan, Ishraq, et al.
Published: (2025)

The Lock-In Phase Hypothesis: Identity Consolidation as a Precursor to AGI
by: Amaral, Marcelo Maciel, et al.
Published: (2025)

SafeAnchor: Preventing Cumulative Safety Erosion in Continual Domain Adaptation of Large Language Models
by: Guo, Dongxin, et al.
Published: (2026)

Unpacking Hateful Memes: Presupposed Context and False Claims
by: Cai, Weibin, et al.
Published: (2025)