paper-with-me

홈 › Papers

Attention in Large Language Models Yields Efficient Zero-Shot Re-Rankers

2024-10-03 · Shijie Chen, Bernal Jiménez Gutiérrez, Yu Su

Information retrieval (IR) systems have played a vital role in modern digital life and have cemented their continued usefulness in this new era of generative AI via retrieval-augmented generation. With strong language processing capabilities and remarkable versatility, large language models (LLMs) have become popular choices for zero-shot re-ranking in IR systems. So far, LLM-based re-ranking methods rely on strong generative capabilities, which restricts their use to either specialized or powerful proprietary models. Given these restrictions, we ask: is autoregressive generation necessary and optimal for LLMs to perform re-ranking? We hypothesize that there are abundant signals relevant to re-ranking within LLMs that might not be used to their full potential via generation. To more directly leverage such signals, we propose in-context re-ranking (ICR), a novel method that leverages the change in attention pattern caused by the search query for accurate and efficient re-ranking. To mitigate the intrinsic biases in LLMs, we propose a calibration method using a content-free query. Due to the absence of generation, ICR only requires two ($O(1)$) forward passes to re-rank $N$ documents, making it substantially more efficient than generative re-ranking methods that require at least $O(N)$ forward passes. Our novel design also enables ICR to be applied to any LLM without specialized training while guaranteeing a well-formed ranking. Extensive experiments with two popular open-weight LLMs on standard single-hop and multi-hop information retrieval benchmarks show that ICR outperforms RankGPT while cutting the latency by more than 60% in practice. Through detailed analyses, we show that ICR's performance is specially strong on tasks that require more complex re-ranking signals. Our findings call for further exploration on novel ways of utilizing open-weight LLMs beyond text generation.

📄 PDF Abstract BibTeX arXiv:2410.02642

Code (0)

등록된 구현이 없습니다.

Tasks

Information RetrievalRe-RankingRetrievalRetrieval-augmented GenerationText Generation

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

Text-Guided Attention is All You Need for Zero-Shot Robustness in Vision-Language Models

2024-10-29 · Lu Yu, Haiyang Zhang, Changsheng Xu

Due to the impressive zero-shot capabilities, pre-trained vision-language models (e.g. CLIP), have attracted widespread attention and adoption across various domains. Nonetheless, CLIP has been observed to be susceptible…

Adversarial RobustnessAll

Zero-Shot Parkinson's Disease Detection from Speech: Comparing Large Audio and Language Models

2026-05-24 · Muhammad Ashad Kabir, Sirajam Munira arxiv

Large audio and language models have recently demonstrated zero-shot reasoning capabilities across various domains. However, it remains unclear how the form of audio input, whether handcrafted acoustic features extracted…

The Power of One: A Single Example is All it Takes for Segmentation in VLMs

2025-03-13 · Mir Rayat Imtiaz Hossain, Mennatullah Siam, Leonid Sigal, James J. Little

Large-scale vision-language models (VLMs), trained on extensive datasets of image-text pairs, exhibit strong multimodal understanding capabilities by implicitly learning associations between textual descriptions and imag…

Allobject-detectionObject DetectionPrompt Engineering+2

Coherence boosting: When your pretrained language model is not paying enough attention

2021-10-15 · ACL 2022 5 · Nikolay Malkin, Zhen Wang, Nebojsa Jojic

Long-range semantic coherence remains a challenge in automatic language generation and understanding. We demonstrate that large language models have insufficiently learned the effect of distant words on next-token predic…

Language ModelingLanguage ModellingText Generation

Humanoid-GPT: Scaling Data and Structure for Zero-Shot Motion Tracking

2026-06-02 · Zekun Qi, Xuchuan Chen, Dairu Liu, Chenghuai Lin 외 arxiv

We introduce Humanoid-GPT, a GPT-style Transformer with causal attention trained on a billion-scale motion corpus for whole-body control. Unlike prior shallow MLP trackers constrained by scarce data and an agility-genera…

Zero-shot Generalization