CORTEXA
← Browse
arxivcs.CLcs.AI2026-06-26

Position Bias Correction is Insufficient for One-Pass Attention Sorting

Qiong Tang, Xiangkun Hu, Xiangyang Liu, Yiran Chen, Yunfan Shao

Long-context language models suffer from position bias, where information in middle positions is underutilized. Attention Sorting addresses this by iteratively reordering documents based on attention patterns, but its multiple sort-and-generate cycles increase deployment cost. We hypothesize that position bias is the primary bottleneck and propose Debiased One-Pass Attention Sorting, which estimates a per-prompt position-bias curve from the low-attention majority of documents and uses it to correct raw attention scores (via subtraction or division) to enable single-pass sorting. Our experiments on two models refute this hypothesis in the tested setting: on LLaMA-2-7B-32K-Instruct, debiasing produces identical results to uncalibrated single-pass sorting (94.83\% containment accuracy), while on YaRN-Llama-2-7b-64k, debiasing improves accuracy by 8.67 percentage points but remains 14.84pp behind iterative sorting, closing only 37\% of the gap. These results suggest that position-bias correction is insufficient to match iterative sorting, and that repeated reordering provides additional benefits beyond bias correction.

View free PDFSource page

Related papers

arxivcs.CLcs.AI2026-07-24

FSE: Continual Learning for Named Entity Recognition by Fast-Slow Experts

Yunan Zhang, Yang Fan, Heng Li, Xiangping Wu, Qingcai Chen

Continual Learning for Named Entity Recognition (CLNER) enable models to incrementally learn new entity types without forgetting previously acquired ones. However, existing methods suffer from catastrophic forgetting and insufficient exploitation of shared information across task…

View free PDFSource page
arxivcs.CLcs.AI2026-07-31

Cross-Lingual Transfer for Machine Translation in Turkic Languages

Omer Burak Cinar, Mehmet Mert Dalkilic, Cagri Toraman

Cross-lingual transfer is central to low-resource machine translation, but its behavior within closely related language families remains insufficiently characterized. We study transfer among five Turkic languages; Turkish, Azerbaijani, Uzbek, Kazakh, and Kyrgyz; using pairwise tr…

View free PDFSource page
arxivcs.CLcs.AIcs.LG2026-07-23

Token Budget Saturation and Mechanistic Early Detection of Reasoning Non-Convergence in Chain-of-Thought Models

Renuka Oladri, Niveda Jawahar, Abdirisak Mohamed

Chain-of-thought reasoning models such as DeepSeek-R1-Distill-Qwen-7B exhibit a bimodal convergence pattern: generations either terminate within a token budget (converged) or exhaust it without reaching a conclusion (non-converged). We characterize this phenomenon empirically, sh…

View free PDFSource page
arxivcs.CLcs.AI2026-07-23

Probing Latent Colombian Identity Inferences in Qwen2.5-7B with Natural Language Autoencoders

Pablo Santiago Potes Velasco, María del Mar García Matabanchoy, Óscar Julián Pérez Ladino, Jhoan Stevan Mosquera Ortiz, Nicolás Lozano Mazuera, Gilber Alexis Corrales Gallego

Large language models may infer demographic attributes from subtle linguistic cues even when those attributes are not explicitly stated. This pilot study examines whether Qwen2.5-7B-Instruct internally represents Colombian identity, socioeconomic status, or stereotype-related inf…

View free PDFSource page
arxivcs.CLcs.AI2026-07-23

Artificial Epanorthosis: Why large language models overuse a classical rhetorical figure, and how to mitigate it

Federico Boggia

A rhetorical figure that Cicero and Quintilian catalogued two thousand years ago reappears, systematically, in the text of large language models: epanorthosis, the self-correction of the specimen «This is not a course. It is a journey of transformation». This essay argues that th…

View free PDFSource page