paper-with-me

Papers

Unable to Forget: Proactive lnterference Reveals Working Memory Limits in LLMs Beyond Context Length

2025-06-09 · Chupei Wang, Jiaqiu Vince Sun

Information retrieval in Large Language Models (LLMs) is increasingly recognized as intertwined with generation capabilities rather than mere lookup. While longer contexts are often assumed to improve retrieval, the effects of intra-context interference remain understudied. To address this, we adapt the proactive interference (PI) paradigm from cognitive science, where earlier information disrupts recall of newer updates. In humans, susceptibility to such interference is inversely linked to working memory capacity. We introduce PI-LLM, an evaluation that sequentially streams semantically related key-value updates and queries only the final values. Although these final values are clearly positioned just before the query, LLM retrieval accuracy declines log-linearly toward zero as interference accumulates; errors arise from retrieving previously overwritten values. Attempts to mitigate interference via prompt engineering (e.g., instructing models to ignore earlier input) yield limited success. These findings reveal a fundamental constraint on LLMs' ability to disentangle interference and flexibly manipulate information, suggesting a working memory bottleneck beyond mere context access. This calls for approaches that strengthen models' ability to suppress irrelevant content during retrieval.

📄 PDF Abstract BibTeX arXiv:2506.08184

Code (0)

등록된 구현이 없습니다.

Tasks

Information RetrievalPrompt EngineeringRetrieval

Similar Papers 제목 키워드 기반

A Novel Framework for Recurrent Neural Networks with Enhancing Information Processing and Transmission between Units

2018-06-02 · Xi Chen, Zhi-Hong Deng, Gehui Shen, Ting Huang

This paper proposes a novel framework for recurrent neural networks (RNNs) inspired by the human memory models in the field of cognitive neuroscience to enhance information processing and transmission between adjacent RN…

General Classificationimage-classificationImage ClassificationLanguage Modelling+2

Compress and Forget: bitsandbytes Quantization Amplifies Proactive Interference in LLMs

2026-08-19 · Shayan Shahrabi-Farahani, Dara Rahmati arxiv

Proactive interference (PI) is a documented failure mode in large language models in which retrieval of a repeatedly overwritten value degrades as prior overwrites accumulate, mirroring a classical phenomenon in human wo…

Convolution Forgetting Curve Model for Repeated Learning

2019-01-19 · Yanlu Xie, Yue Chen, Man Li

Most of mathematic forgetting curve models fit well with the forgetting data under the learning condition of one time rather than repeated. In the paper, a convolution model of forgetting curve is proposed to simulate th…

model

SIA: Symbolic Interpretability for Anticipatory Deep Reinforcement Learning in Network Control

2026-01-29 · MohammadErfan Jabbari, Abhishek Duttagupta, Claudio Fiandrino, Leonardo Bonati 외 arxiv

Deep reinforcement learning (DRL) promises adaptive control for future mobile networks but conventional agents remain reactive: they act on past and current measurements and cannot leverage short-term forecasts of exogen…

Reinforcement LearningKnowledge Graphs

AtlasVLA: Persistent World-Ego State Modeling for Vision-Language-Action Models

2026-08-07 · Guiyu Zhao, Longteng Guo, Yanghong Mei, Zilin Zhu 외 hf

While Vision-Language-Action (VLA) models have advanced embodied AI, their fundamentally reactive paradigm severely limits performance in partially observable and long-horizon tasks. When restricted to a single wrist-mou…

Spatial Reasoning