Authored Publications
Sort By
PRewrite: Prompt Rewriting with Reinforcement Learning
Qiaozhu Mei
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (2024) (to appear)
Preview abstract
Prompt engineering is critical for the development of LLM-based applications. However, it is usually done manually in a "trial and error" fashion that can be time consuming, ineffective, and sub-optimal. Even for the prompts which seemingly work well, there is always a lingering question: can the prompts be made better with further modifications?
To address these problems, we investigate automated prompt engineering in this paper. Specifically, we propose PRewrite, an automated method to rewrite an under-optimized prompt to a more effective prompt. We instantiate the prompt rewriter using an LLM. The rewriter LLM is trained using reinforcement learning to optimize the performance on a given downstream task. We conduct experiments on diverse benchmark datasets, which demonstrates the effectiveness of PRewrite.
View details
Knowledge Distillation with Perturbed Loss: From a Vanilla Teacher to a Proxy Teacher
Rongzhi Zhang
Chao Zhang
Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2024), ACM, pp. 4278 - 4289
Preview abstract
Knowledge distillation is a popular technique to transfer knowledge from a large teacher model to a small student model. Typically, the student learns to imitate the teacher by minimizing the KL divergence of its output distribution with the teacher's output distribution. In this work, we argue that such a learning objective is sub-optimal because there exists a discrepancy between the teacher's output distribution and the ground truth label distribution. Therefore, forcing the student to blindly imitate the unreliable teacher output distribution leads to inferior performance. To this end, we propose a novel knowledge distillation objective PTLoss by first representing the vanilla KL-based distillation loss function via a Maclaurin series and then perturbing the leading-order terms in this series. This perturbed loss implicitly transforms the original teacher into a proxy teacher with a distribution closer to the ground truth distribution. We establish the theoretical connection between this "distribution closeness'' and the student model generalizability, which enables us to select the PTLoss's perturbation coefficients in a principled way. Extensive experiments on six public benchmark datasets demonstrate the effectiveness of PTLoss with teachers of different scales.
View details
Beyond Yes and No: Improving Zero-Shot Pointwise LLM Rankers via Scoring Fine-Grained Relevance Labels
Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL)
Preview abstract
Zero-shot text rankers powered by recent LLMs achieve remarkable ranking performance by simply prompting. Existing prompts for pointwise LLM rankers mostly ask the model to choose from binary relevance labels like "Yes" and "No". However, the lack of intermediate relevance label options may cause the LLM to provide noisy or biased answers for documents that are partially relevant to the query. We propose to incorporate fine-grained relevance labels into the prompt for LLM rankers, enabling them to better differentiate among documents with different levels of relevance to the query and thus derive a more accurate ranking. We study two variants of the prompt template, coupled with different numbers of relevance levels. Our experiments on 8 BEIR data sets show that adding fine-grained relevance labels significantly improves the performance of LLM rankers.
View details
Learning to Rewrite Prompts for Personalized Text Generation
Qiaozhu Mei
Proceedings of the ACM Web Conference 2024
Preview abstract
Facilitated by large language models (LLMs), personalized text generation has become a rapidly growing research direction. Most existing studies focus on designing specialized models for a particular domain, or they require fine-tuning the LLMs to generate personalized text. We consider a typical scenario in which the large language model, which generates personalized output, is frozen and can only be accessed through APIs. Under this constraint, all one can do is to improve the input text (i.e., text prompts) sent to the LLM, a procedure that is usually done manually. In this paper, we propose a novel method to automatically revise prompts for personalized text generation. The proposed method takes the initial prompts generated by a state-of-the-art, multistage framework for personalized generation and rewrites a few critical components that summarize and synthesize the personal context. The prompt rewriter employs a training paradigm that chains together supervised learning (SL) and reinforcement learning (RL), where SL reduces the search space of RL and RL facilitates end-to-end training of the rewriter. Using datasets from three representative domains, we demonstrate that the rewritten prompts outperform both the original prompts and the prompts optimized via supervised learning or reinforcement learning alone. In-depth analysis of the rewritten prompts shows that they are not only human readable, but also able to guide manual revision of prompts when there is limited resource to employ reinforcement learning to train the prompt rewriter, or when it is costly to deploy an automatic prompt rewriter for inference.
View details
Preview abstract
In-Context Learning (ICL) is an emergent capability of Large Language Models (LLMs).
Only a few demonstrations enable LLMs to be used as blackbox for new tasks. Previous studies have shown that using LLMs' outputs as labels is effective in training models to select demonstrations. Such a label is expected to estimate utility of a demonstration in ICL;
however, it has not been well understood how different labeling strategies affect results on target tasks. This paper presents an analysis on different utility functions by focusing on LLMs' output probability given ground-truth output, and task-specific reward given LLMs' prediction. Unlike the previous work, we introduce a novel labeling method, incremental utility, which estimates how much incremental knowledge is brought into the LLMs by a demonstration. We conduct experiments with instruction-tuned LLMs on binary/multi-class classification, segmentation, and translation across Arabic, English, Finnish, Japanese, and Spanish. Our results show that (1) the probability is effective when the probability values are distributed across the whole value range (on the classification tasks), and (2) the downstream metric is more robust when nuanced reward values are provided with long outputs (on the segmentation and translation tasks). We then show that the proposed incremental utility further helps ICL by contrasting how the LLMs perform with and without the demonstrations.
View details
Bridging the Preference Gap between Retrievers and LLMs
Zixuan Ke
Qiaozhu Mei
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (2024) (to appear)
Preview abstract
Large Language Models (LLMs) have demonstrated superior results across a wide range of tasks, and Retrieval-augmented Generation (RAG) is an effective way to enhance the performance by locating relevant information and placing it into the context window of the LLM. However, the relationship between retrievers and LLM in a RAG is still under-investigated. Most existing work treats the retriever and the LLM as independent components and leaves a gap between retrieving human-"friendly" information and assembling a LLM-"friendly" context. In this work, we examine a novel bridge mechanism. We validate the ranking and selection assumptions of retrievers in the context of RAG and propose a framework that chains together supervised and reinforcement learning to train a bridge model that optimizes the connection between the retriever and the LLM. Empirical results demonstrate the effectiveness of our method in both question-answering and personalized generation tasks.
View details
Large Language Models are Effective Text Rankers with Pairwise Ranking Prompting
Conference of the North American Chapter of the Association for Computational Linguistics (NAACL) (2024)
Preview abstract
Ranking documents using Large Language Models (LLMs) by directly feeding the query and candidate documents into the prompt is an interesting and practical problem. However, researchers have found it difficult to outperform fine-tuned baseline rankers on benchmark datasets. We analyze pointwise and listwise ranking prompts used by existing methods and argue that off-the-shelf LLMs do not fully understand these challenging ranking formulations. In this paper, we propose to significantly reduce the burden on LLMs by using a new technique called Pairwise Ranking Prompting (PRP). Our results are the first in the literature to achieve state-of-the-art ranking performance on standard benchmarks using moderate-sized open-sourced LLMs. On TREC-DL 2019&2020, PRP based on the Flan-UL2 model with 20B parameters performs favorably with the previous best approach in the literature, which is based on the blackbox commercial GPT-4 that has 50x (estimated) model size, while outperforming other LLM-based solutions, such as InstructGPT which has 175B parameters, by over 10% for all ranking metrics. By using the same prompt template on seven BEIR tasks, PRP outperforms supervised baselines and outperforms the blackbox commercial ChatGPT solution by 4.2% and pointwise LLM-based solutions by more than 10% on average NDCG@10. Furthermore, we propose several variants of PRP to improve efficiency and show that it is possible to achieve competitive results even with linear complexity.
View details
Can Query Expansion Improve Generalization of Strong Cross-Encoder Rankers?
Minghan Li
Jimmy Lin
Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’24) (2024)
Preview abstract
Query expansion has been widely used to improve the search results of first-stage retrievers, yet its influence on second-stage, crossencoder rankers remains under-explored. A recent study shows that current expansion techniques benefit weaker models but harm stronger rankers. In this paper, we re-examine this conclusion and raise the following question: Can query expansion improve generalization of strong cross-encoder rankers? To answer this question, we first apply popular query expansion methods to different crossencoder rankers and verify the deteriorated zero-shot effectiveness. We identify two vital steps in the experiment: high-quality keyword generation and minimally-disruptive query modification. We show that it is possible to improve the generalization of a strong neural ranker, by generating keywords through a reasoning chain and aggregating the ranking results of each expanded query via selfconsistency, reciprocal rank weighting, and fusion. Experiments on BEIR and TREC Deep Learning 2019/2020 show that the nDCG@10 scores of both MonoT5 and RankT5 following these steps are improved, which points out a direction for applying query expansion to strong cross-encoder rankers.
View details
WSDM 2024 Workshop on Large Language Models for Individuals, Groups, and Society
Qiaozhu Mei
Vanessa Murdock
Jie Tang
Hongning Wang
Hamed Zamani
(2024) (to appear)
Preview abstract
This workshop discusses the cutting-edge developments in research and applications of personalizing large language models (LLMs) and adapting them to the demands of diverse user populations and societal needs. The full-day workshop plan includes several keynotes and invited talks, a poster session and a panel discussion.
View details
Its All Relative! -- A Synthetic Query Generation Approach for Improving Zero-Shot Relevance Prediction
Findings of the Association for Computational Linguistics: NAACL 2024
Preview abstract
Recent developments in large language models (LLMs) have shown promise in their ability to generate synthetic query-document pairs by prompting LLMs with as few as 8 demonstrations \cite{dai2022promptagator}.
This has enabled building better IR models especially for tasks which have no training data readily available.
Typically, such synthetic query generation (QGen) approaches condition on an input context (e.g. document) and generate a query that is relevant to that context or condition the QGen model additionally on the relevance label (e.g. relevant vs irrelevant) to generate queries across relevance buckets.
However, we find that such QGen approaches are sub-optimal as it requires the model to reason about the desired label and the input from only a handful of examples, which is not trivial, especially when the relevance buckets are nuanced.
In this work, we propose to reduce this burden of LLMs by generating queries simultaneously for different labels (e.g. relevance buckets).
We hypothesize that instead of asking the model to generate, say, an irrelevant query given an input context, asking the model to generate an irrelevant query with respect to a relevant query is a much simpler task setup for the model to reason about.
Extensive experimentation across seven IR datasets shows that synthetic queries generated in such a fashion translates to a better downstream performance, suggesting that the generated queries are indeed of higher quality.
View details
