Variations in Relevance Judgments and the Shelf Life of Test Collections

The fundamental property of Cranfield-style evaluations, that system rankings are stable even when assessors disagree on individual relevance decisions, was validated on traditional test collections. However, the paradigm shift towards neural retrieval models affected the characteristics of modern test collections, e.g., documents are short, judged with four grades of relevance, and information needs have no descriptions or narratives. Under these changes, it is unclear whether assessor disagreement remains negligible for system comparisons. We investigate this aspect under the additional condition that the few modern test collections are heavily re-used. Given more possible query interpretations due to less formalized information needs, an ''expiration date'' for test collections might be needed if top-effectiveness requires overfitting to a single interpretation of relevance. We run a reproducibility study and re-annotate the relevance judgments of the 2019 TREC Deep Learning track. We can reproduce prior work in the neural retrieval setting, showing that assessor disagreement does not affect system rankings. However, we observe that some models substantially degrade with our new relevance judgments, and some have already reached the effectiveness of humans as rankers, providing evidence that test collections can expire.

DenseReviewer: A Screening Prioritisation Tool for Systematic Review based on Dense Retrieval

Screening is a time-consuming and labour-intensive yet required task for medical systematic reviews, as tens of thousands of studies often need to be screened. Prioritising relevant studies to be screened allows downstream systematic review creation tasks to start earlier and save time. In previous work, we developed a dense retrieval method to prioritise relevant studies with reviewer feedback during the title and abstract screening stage. Our method outperforms previous active learning methods in both effectiveness and efficiency. In this demo, we extend this prior work by creating (1) a web-based screening tool that enables end-users to screen studies exploiting state-of-the-art methods and (2) a Python library that integrates models and feedback mechanisms and allows researchers to develop and demonstrate new active learning methods. We describe the tool's design and showcase how it can aid screening. The tool is available at https://densereviewer.ielab.io. The source code is also open sourced at https://github.com/ielab/densereviewer.

Ranking Generated Answers: On the Agreement of Retrieval Models with Humans on Consumer Health Questions

Evaluating the output of generative large language models (LLMs) is challenging and difficult to scale. Most evaluations of LLMs focus on tasks such as single-choice question-answering or text classification. These tasks are not suitable for assessing open-ended question-answering capabilities, which are critical in domains where expertise is required, such as health, and where misleading or incorrect answers can have a significant impact on a user's health. Using human experts to evaluate the quality of LLM answers is generally considered the gold standard, but expert annotation is costly and slow. We present a method for evaluating LLM answers that uses ranking signals as a substitute for explicit relevance judgements. Our scoring method correlates with the preferences of human experts. We validate it by investigating the well-known fact that the quality of generated answers improves with the size of the model as well as with more sophisticated prompting strategies.

Learning Effective Representations for Retrieval Using Self-Distillation with Adaptive Relevance Margins

Representation-based retrieval models, so-called biencoders, estimate the relevance of a document to a query by calculating the similarity of their respective embeddings. Current state-of-the-art biencoders are trained using an expensive training regime involving knowledge distillation from a teacher model and batch-sampling. Instead of relying on a teacher model, we contribute a novel parameter-free loss function for self-supervision that exploits the pre-trained language modeling capabilities of the encoder model as a training signal, eliminating the need for batch sampling by performing implicit hard negative mining. We investigate the capabilities of our proposed approach through extensive ablation studies, demonstrating that self-distillation can match the effectiveness of teacher distillation using only 13.5% of the data, while offering a speedup in training time between 3x and 15x compared to parametrized losses. Code and data is made openly available.

A Systematic Investigation of Distilling Large Language Models into Cross-Encoders for Passage Re-ranking

Cross-encoders distilled from large language models are more effective re-rankers than cross-encoders fine-tuned using manually labeled data. However, the distilled models do not reach the language model's effectiveness. We construct and release a new distillation dataset, named Rank-DistiLLM, to investigate whether insights from fine-tuning cross-encoders on manually labeled data -- hard-negative sampling, deep sampling, and listwise loss functions -- are transferable to large language model ranker distillation. Our dataset can be used to train cross-encoders that reach the effectiveness of large language models while being orders of magnitude more efficient. Code and data is available at: https://github.com/webis-de/msmarco-llm-distillation

Set-Encoder: Permutation-Invariant Inter-Passage Attention for Listwise Passage Re-Ranking with Cross-Encoders

Cross-encoders are effective passage re-rankers. But when re-ranking multiple passages at once, existing cross-encoders inefficiently optimize the output ranking over several input permutations, as their passage interactions are not permutation-invariant. Moreover, their high memory footprint constrains the number of passages during listwise training. To tackle these issues, we propose the Set-Encoder, a new cross-encoder architecture that (1) introduces inter-passage attention with parallel passage processing to ensure permutation invariance between input passages, and that (2) uses fused-attention kernels to enable training with more passages at a time. In experiments on TREC Deep Learning and TIREx, the Set-Encoder is more effective than previous cross-encoders with a similar number of parameters. Compared to larger models, the Set-Encoder is more efficient and either on par or even more effective.

Zero-shot Generative Large Language Models for Systematic Review Screening Automation

Systematic reviews are crucial for evidence-based medicine as they comprehensively analyse published research findings on specific questions. Conducting such reviews is often resource- and time-intensive, especially in the screening phase, where abstracts of publications are assessed for inclusion in a review. This study investigates the effectiveness of using zero-shot large language models~(LLMs) for automatic screening. We evaluate the effectiveness of eight different LLMs and investigate a calibration technique that uses a predefined recall threshold to determine whether a publication should be included in a systematic review. Our comprehensive evaluation using five standard test collections shows that instruction fine-tuning plays an important role in screening, that calibration renders LLMs practical for achieving a targeted recall, and that combining both with an ensemble of zero-shot models saves significant screening time compared to state-of-the-art approaches.

Evaluating Generative Ad Hoc Information Retrieval

Recent advances in large language models have enabled the development of viable generative information retrieval systems. A generative retrieval system returns a grounded generated text in response to an information need instead of the traditional document ranking. Quantifying the utility of these types of responses is essential for evaluating generative retrieval systems. As the established evaluation methodology for ranking-based ad hoc retrieval may seem unsuitable for generative retrieval, new approaches for reliable, repeatable, and reproducible experimentation are required. In this paper, we survey the relevant information retrieval and natural language processing literature, identify search tasks and system architectures in generative retrieval, develop a corresponding user model, and study its operationalization. This theoretical analysis provides a foundation and new insights for the evaluation of generative ad hoc retrieval systems.

Generating Natural Language Queries for More Effective Systematic Review Screening Prioritisation

Screening prioritisation in medical systematic reviews aims to rank the set of documents retrieved by complex Boolean queries. The goal is to prioritise the most important documents so that subsequent review steps can be carried out more efficiently and effectively. The current state of the art uses the final title of the review to rank documents using BERT-based neural neural rankers. However, the final title is only formulated at the end of the review process, which makes this approach impractical as it relies on ex post facto information. At the time of screening, only a rough working title is available, with which the BERT-based ranker achieves is significantly worse than the final title. In this paper, we explore alternative sources of queries for screening prioritisation, such as the Boolean query used to retrieve the set of documents to be screened, and queries generated by instruction-based generative large language models such as ChatGPT and Alpaca. Our best approach is not only practical based on the information available at screening time, but is similar in effectiveness with the final title.

Beyond CO2 Emissions: The Overlooked Impact of Water Consumption of Information Retrieval Models

As in other fields of artificial intelligence, the information retrieval community has grown interested in investigating the power consumption associated with neural models, particularly models of search. This interest has become particularly relevant as the energy consumption of information retrieval models has risen with new neural models based on large language models, leading to an associated increase of CO2 emissions, albeit relatively low compared to fields such as natural language processing.