paper-with-me

홈 › Papers

Same Task, More Tokens: the Impact of Input Length on the Reasoning Performance of Large Language Models

2024-02-19 · Mosh Levy, Alon Jacoby, Yoav Goldberg

This paper explores the impact of extending input lengths on the capabilities of Large Language Models (LLMs). Despite LLMs advancements in recent times, their performance consistency across different input lengths is not well understood. We investigate this aspect by introducing a novel QA reasoning framework, specifically designed to assess the impact of input length. We isolate the effect of input length using multiple versions of the same sample, each being extended with padding of different lengths, types and locations. Our findings show a notable degradation in LLMs' reasoning performance at much shorter input lengths than their technical maximum. We show that the degradation trend appears in every version of our dataset, although at different intensities. Additionally, our study reveals that the traditional metric of next word prediction correlates negatively with performance of LLMs' on our reasoning dataset. We analyse our results and identify failure modes that can serve as useful guides for future research, potentially informing strategies to address the limitations observed in LLMs.

📄 PDF Abstract BibTeX arXiv:2402.14848

Code (1)

alonj/Same-Task-More-Tokens 공식 구현

Similar Papers 제목 키워드 기반

Vocabulary Customization for Efficient Domain-Specific LLM Deployment

2025-09-30 · Christian Herold, Michael Kozielski, Nicholas Santavas, Yannick Versley 외 arxiv

When using an LLM to process text outside the training domain(s), an often overlooked factor is vocabulary mismatch, where the general-domain tokenizer fails to capture frequent domain-specific terms, leading to higher t…

How can objects help action recognition?

2023-06-20 · CVPR 2023 1 · Xingyi Zhou, Anurag Arnab, Chen Sun, Cordelia Schmid

Current state-of-the-art video models process a video clip as a long sequence of spatio-temporal tokens. However, they do not explicitly model objects, their interactions across the video, and instead process all the tok…

Action RecognitionObject

ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators

2020-03-23 · ICLR 2020 1 · Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning

Masked language modeling (MLM) pre-training methods such as BERT corrupt the input by replacing some tokens with [MASK] and then train a model to reconstruct the original tokens. While they produce good results when tran…

GPULanguage ModelingLanguage ModellingMasked Language Modeling+3

On Sensitivity of Deep Learning Based Text Classification Algorithms to Practical Input Perturbations

2022-01-02 · Aamir Miyajiwala, Arnav Ladkat, Samiksha Jagadale, Raviraj Joshi

Text classification is a fundamental Natural Language Processing task that has a wide variety of applications, where deep learning approaches have produced state-of-the-art results. While these models have been heavily c…

Deep LearningSensitivitytext-classificationText Classification

Explaining Why: How Instructions and User Interfaces Impact Annotator Rationales When Labeling Text Data

2022-07-01 · NAACL 2022 7 · Cynthia Sullivan, William Brackenbury, Andrew McNut, Kevin Bryson 외

In the context of data labeling, NLP researchers are increasingly interested in having humans select rationales, a subset of input tokens relevant to the chosen label. We conducted a 332-participant online user study to …