Copertina

When LLMs Stop Following Steps: A Diagnostic Study of Procedural Execution in Language Models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Panda, Sailesh, Kadasi, Pritam, Upperwal, Abhishek, Singh, Mayank
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Task--Specificity Score: Measuring How Much Instructions Really Matter for Supervision
di: Kadasi, Pritam, et al.
Pubblicazione: (2026)

ADAPT: Learning Task Mixtures for Budget-Constrained Instruction Tuning
di: Kadasi, Pritam, et al.
Pubblicazione: (2025)

Eka-Eval: An Evaluation Framework for Low-Resource Multilingual Large Language Models
di: Sinha, Samridhi Raj, et al.
Pubblicazione: (2025)

Char-mander Use mBackdoor! A Study of Cross-lingual Backdoor Attacks in Multilingual LLMs
di: Beniwal, Himanshu, et al.
Pubblicazione: (2025)

Model Hubs and Beyond: Analyzing Model Popularity, Performance, and Documentation
di: Kadasi, Pritam, et al.
Pubblicazione: (2025)

Evaluating LLMs' Reasoning Over Ordered Procedural Steps
di: Anika, Adrita, et al.
Pubblicazione: (2025)

Know When To Stop: A Study of Semantic Drift in Text Generation
di: Spataru, Ava, et al.
Pubblicazione: (2024)

Unmasking Hallucinations: A Causal Graph-Attention Perspective on Factual Reliability in Large Language Models
di: kurra, Sailesh kiran, et al.
Pubblicazione: (2026)

ProcBench: Benchmark for Multi-Step Reasoning and Following Procedure
di: Fujisawa, Ippei, et al.
Pubblicazione: (2024)

Stop When Enough: Adaptive Early-Stopping for Chain-of-Thought Reasoning
di: Sun, Renliang, et al.
Pubblicazione: (2025)

L0-Reasoning Bench: Evaluating Procedural Correctness in Language Models via Simple Program Execution
di: Sun, Simeng, et al.
Pubblicazione: (2025)

Stop Rewarding Hallucinated Steps: Faithfulness-Aware Step-Level Reinforcement Learning for Small Reasoning Models
di: Nie, Shuo, et al.
Pubblicazione: (2026)

Where Does Toxicity Live? Mechanistic Localization and Targeted Suppression in Language Models
di: Beniwal, Himanshu, et al.
Pubblicazione: (2026)

When Thinking Fails: The Pitfalls of Reasoning for Instruction-Following in LLMs
di: Li, Xiaomin, et al.
Pubblicazione: (2025)

SmoGVLM: A Small, Graph-enhanced Vision-Language Model
di: Mondal, Debjyoti, et al.
Pubblicazione: (2026)

PythonSaga: Redefining the Benchmark to Evaluate Code Generating LLMs
di: Yadav, Ankit, et al.
Pubblicazione: (2024)

Step-by-Step Mastery: Enhancing Soft Constraint Following Ability of Large Language Models
di: Ren, Qingyu, et al.
Pubblicazione: (2025)

Cross-lingual Editing in Multilingual Language Models
di: Beniwal, Himanshu, et al.
Pubblicazione: (2024)

Stop When Reasoning Converges: Semantic-Preserving Early Exit for Reasoning Models
di: Min, Dehai, et al.
Pubblicazione: (2026)

Mitigating the Impact of Outlier Channels for Language Model Quantization with Activation Regularization
di: Nrusimha, Aniruddha, et al.
Pubblicazione: (2024)

A Call for Clarity in Beam Search: How It Works and When It Stops
di: Kasai, Jungo, et al.
Pubblicazione: (2022)

Joint Action Language Modelling for Transparent Policy Execution
di: Wulff, Theodor, et al.
Pubblicazione: (2025)

Thinking Out of Order: When Output Order Stops Reflecting Reasoning Order in Diffusion Language Models
di: Yu, Longxuan, et al.
Pubblicazione: (2026)

TRACES: Tagging Reasoning Steps for Adaptive Cost-Efficient Early-Stopping
di: Belkhiter, Yannis, et al.
Pubblicazione: (2026)

The Model's Language Matters: A Comparative Privacy Analysis of LLMs
di: Mishra, Abhishek K., et al.
Pubblicazione: (2025)

Early Stopping Chain-of-thoughts in Large Language Models
di: Mao, Minjia, et al.
Pubblicazione: (2025)

When Context Leads but Parametric Memory Follows in Large Language Models
di: Tao, Yufei, et al.
Pubblicazione: (2024)

How Robust are the Tabular QA Models for Scientific Tables? A Study using Customized Dataset
di: Ghosh, Akash, et al.
Pubblicazione: (2024)

When to Memorize and When to Stop: Gated Recurrent Memory for Long-Context Reasoning
di: Sheng, Leheng, et al.
Pubblicazione: (2026)

Learning When to Act or Refuse: Guarding Agentic Reasoning Models for Safe Multi-Step Tool Use
di: Agarwal, Aradhye, et al.
Pubblicazione: (2026)

SOPBench: Evaluating Language Agents at Following Standard Operating Procedures and Constraints
di: Li, Zekun, et al.
Pubblicazione: (2025)

Beyond Monolingual Assumptions: A Survey of Code-Switched NLP in the Era of Large Language Models across Modalities
di: Sheth, Rajvee, et al.
Pubblicazione: (2025)

Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models
di: Sui, Yang, et al.
Pubblicazione: (2025)

How Linguistics Learned to Stop Worrying and Love the Language Models
di: Futrell, Richard, et al.
Pubblicazione: (2025)

Strong Memory, Weak Control: An Empirical Study of Executive Functioning in LLMs
di: de Langis, Karin, et al.
Pubblicazione: (2025)

Defining and Evaluating Decision and Composite Risk in Language Models Applied to Natural Language Inference
di: Shen, Ke, et al.
Pubblicazione: (2024)

Knowing When to Stop: Efficient Context Processing via Latent Sufficiency Signals
di: Xie, Roy, et al.
Pubblicazione: (2025)

Reasoning or a Semblance of it? A Diagnostic Study of Transitive Reasoning in LLMs
di: Mehrafarin, Houman, et al.
Pubblicazione: (2024)

One Instruction Does Not Fit All: How Well Do Embeddings Align Personas and Instructions in Low-Resource Indian Languages?
di: Shah, Arya, et al.
Pubblicazione: (2026)

When LLMs Meet Cunning Texts: A Fallacy Understanding Benchmark for Large Language Models
di: Li, Yinghui, et al.
Pubblicazione: (2024)