Cover Image

Positional Failures in Long-Context LLMs: A Blind Spot in Reasoning Benchmarks

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Chuyifei, Cui, Hongyu, Huang, Xiaowen, Sang, Jitao
Format:	Preprint
Published:	2026
Subjects:	Computation and Language Artificial Intelligence Machine Learning I.2.7
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

IdentifyMe: A Challenging Long-Context Mention Resolution Benchmark for LLMs
by: Manikantan, Kawshik, et al.
Published: (2024)

Below-Chance Blindness: Prompted Underperformance in Small LLMs Produces Positional Bias Rather than Answer Avoidance
by: Cacioli, Jon-Paul
Published: (2026)

RomanLens: The Role Of Latent Romanization In Multilinguality In LLMs
by: Saji, Alan, et al.
Published: (2025)

Evaluating Long Range Dependency Handling in Code Generation LLMs
by: Assogba, Yannick, et al.
Published: (2024)

Evaluating the Efficacy of Hybrid Deep Learning Models in Distinguishing AI-Generated Text
by: Oketunji, Abiodun Finbarrs
Published: (2023)

Contextual Integrity in LLMs via Reasoning and Reinforcement Learning
by: Lan, Guangchen, et al.
Published: (2025)

KV-Fold: One-Step KV-Cache Recurrence for Long-Context Inference
by: Nadali, Alireza, et al.
Published: (2026)

Efficient Solutions For An Intriguing Failure of LLMs: Long Context Window Does Not Mean LLMs Can Analyze Long Sequences Flawlessly
by: Hosseini, Peyman, et al.
Published: (2024)

From Benchmarking to Reasoning: A Dual-Aspect, Large-Scale Evaluation of LLMs on Vietnamese Legal Text
by: Le, Van-Truong
Published: (2026)

Mathador-LM: A Dynamic Benchmark for Mathematical Reasoning on Large Language Models
by: Kurtic, Eldar, et al.
Published: (2024)

ALBA: A European Portuguese Benchmark for Evaluating Language and Linguistic Dimensions in Generative LLMs
by: Vieira, Inês, et al.
Published: (2026)

Detecting Data Contamination in LLMs via In-Context Learning
by: Zawalski, Michał, et al.
Published: (2025)

Context Is What You Need: The Maximum Effective Context Window for Real World Limits of LLMs
by: Paulsen, Norman
Published: (2025)

Cultural Benchmarking of LLMs in Standard and Dialectal Arabic Dialogues
by: Kautsar, Muhammad Dehan Al, et al.
Published: (2026)

Cognitive Load Limits in Large Language Models: Benchmarking Multi-Hop Reasoning
by: Adapala, Sai Teja Reddy
Published: (2025)

Text-Based Approaches to Item Difficulty Modeling in Large-Scale Assessments: A Systematic Review
by: Peters, Sydney, et al.
Published: (2025)

Measuring Reasoning Utility in LLMs via Conditional Entropy Reduction
by: Guo, Xu
Published: (2025)

Benchmarking the Performance of Pre-trained LLMs across Urdu NLP Tasks
by: Tahir, Munief Hassan, et al.
Published: (2024)

UA-Legal-Bench: A Benchmark for Evaluating Large Language Models on Ukrainian Legal Reasoning
by: Ovcharov, Volodymyr
Published: (2026)

Self-Supervised Position Debiasing for Large Language Models
by: Liu, Zhongkun, et al.
Published: (2024)

Syntactic Blind Spots: How Misalignment Leads to LLMs Mathematical Errors
by: Williamson, Dane, et al.
Published: (2025)

KisMATH: Do LLMs Have Knowledge of Implicit Structures in Mathematical Reasoning?
by: Saha, Soumadeep, et al.
Published: (2025)

Does LLM Alignment Really Need Diversity? An Empirical Study of Adapting RLVR Methods for Moral Reasoning
by: Zhang, Zhaowei, et al.
Published: (2026)

Blind Spots in the Guard: How Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems
by: Pai, Aaditya
Published: (2026)

Enhancing In-Context Learning via Implicit Demonstration Augmentation
by: Zhou, Xiaoling, et al.
Published: (2024)

Adaptive Interviewing for Persona Simulation in LLMs: Evidence-Grounded Reasoning Improves Decision Alignment
by: Su, Ruoxi, et al.
Published: (2026)

GMoE: Empowering LLMs Fine-Tuning via MoE Graph Collaboration
by: Bai, Ting, et al.
Published: (2024)

MORABLES: A Benchmark for Assessing Abstract Moral Reasoning in LLMs with Fables
by: Marcuzzo, Matteo, et al.
Published: (2025)

EduAdapt: A Question Answer Benchmark Dataset for Evaluating Grade-Level Adaptability in LLMs
by: Naeem, Numaan, et al.
Published: (2025)

Cetvel: A Unified Benchmark for Evaluating Language Understanding, Generation and Cultural Capacity of LLMs for Turkish
by: Er, Yakup Abrek, et al.
Published: (2025)

Can AI Read Between The Lines? Benchmarking LLMs On Financial Nuance
by: Kubica, Dominick, et al.
Published: (2025)

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders
by: Wu, Zhengxuan, et al.
Published: (2025)

LongDocFACTScore: Evaluating the Factuality of Long Document Abstractive Summarisation
by: Bishop, Jennifer A, et al.
Published: (2023)

Language Models are Hidden Reasoners: Unlocking Latent Reasoning Capabilities via Self-Rewarding
by: Chen, Haolin, et al.
Published: (2024)

RAVR: Reference-Answer-guided Variational Reasoning for Large Language Models
by: Lin, Tianqianjin, et al.
Published: (2025)

RAudit: A Blind Auditing Protocol for Large Language Model Reasoning
by: Chang, Edward Y., et al.
Published: (2026)

UrduBench: An Urdu Reasoning Benchmark using Contextually Ensembled Translations with Human-in-the-Loop
by: Shafique, Muhammad Ali, et al.
Published: (2026)

Large Language Model (LLM) Bias Index -- LLMBI
by: Oketunji, Abiodun Finbarrs, et al.
Published: (2023)

The Metacognitive Probe: Five Behavioural Calibration Diagnostics for LLMs
by: Oliveira, Rafael C. T.
Published: (2026)

IntentGrasp: A Comprehensive Benchmark for Intent Understanding
by: Yin, Yuwei, et al.
Published: (2026)