Collapse of Dense Retrievers: Short, Early, and Literal Biases Outranking Factual Evidence

• URL: http://arxiv.org/abs/2503.05037v2
• Date: Mon, 02 Jun 2025 22:19:43 GMT
• Title: Collapse of Dense Retrievers: Short, Early, and Literal Biases Outranking Factual Evidence
• Authors: Mohsen Fayyaz, Ali Modarressi, Hinrich Schuetze, Nanyun Peng,
• Abstract summary: Retrieval models are commonly used in Information Retrieval (IR) applications, such as Retrieval-Augmented Generation (RAG)<n>We quantify the impact of biases, such as a preference for shorter documents, on retrievers like Dragon+ and Contriever.<n>We uncover major vulnerabilities, showing retrievers favor shorter documents, early positions, repeated entities, and literal matches, all while ignoring the answer's presence!
• Score: 56.09494651178128
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Dense retrieval models are commonly used in Information Retrieval (IR) applications, such as Retrieval-Augmented Generation (RAG). Since they often serve as the first step in these systems, their robustness is critical to avoid downstream failures. In this work, we repurpose a relation extraction dataset (e.g., Re-DocRED) to design controlled experiments that quantify the impact of heuristic biases, such as a preference for shorter documents, on retrievers like Dragon+ and Contriever. We uncover major vulnerabilities, showing retrievers favor shorter documents, early positions, repeated entities, and literal matches, all while ignoring the answer's presence! Notably, when multiple biases combine, models exhibit catastrophic performance degradation, selecting the answer-containing document in less than 10% of cases over a synthetic biased document without the answer. Furthermore, we show that these biases have direct consequences for downstream applications like RAG, where retrieval-preferred documents can mislead LLMs, resulting in a 34% performance drop than providing no documents at all. https://huggingface.co/datasets/mohsenfayyaz/ColDeR

Related papers

• Learning More Effective Representations for Dense Retrieval through Deliberate Thinking Before Search [65.53881294642451]
  Deliberate Thinking based Dense Retriever (DEBATER) DEBATER enhances recent dense retrievers by enabling them to learn more effective document representations through a step-by-step thinking process. Experimental results show that DEBATER significantly outperforms existing methods across several retrieval benchmarks.
  arXiv  Detail & Related papers  (2025-02-18T15:56:34Z)
• Riddle Me This! Stealthy Membership Inference for Retrieval-Augmented Generation [18.098228823748617]
  We present Interrogation Attack (IA), a membership inference technique targeting documents in the RAG datastore.<n>We demonstrate successful inference with just 30 queries while remaining stealthy.<n>We observe a 2x improvement in TPR@1%FPR over prior inference attacks across diverse RAG configurations.
  arXiv  Detail & Related papers  (2025-02-01T04:01:18Z)
• Drowning in Documents: Consequences of Scaling Reranker Inference [35.499018267073964]
  Cross-encoders are often used to re-score the documents retrieved by cheaper initial IR systems. We measure reranker performance for full retrieval, not just re-scoring first-stage retrieval. Our experiments reveal a surprising trend: the best existing rerankers provide diminishing returns when scoring progressively more documents.
  arXiv  Detail & Related papers  (2024-11-18T17:46:32Z)
• ExcluIR: Exclusionary Neural Information Retrieval [74.08276741093317]
  We present ExcluIR, a set of resources for exclusionary retrieval. evaluation benchmark includes 3,452 high-quality exclusionary queries. training set contains 70,293 exclusionary queries, each paired with a positive document and a negative document.
  arXiv  Detail & Related papers  (2024-04-26T09:43:40Z)
• List-aware Reranking-Truncation Joint Model for Search and Retrieval-augmented Generation [80.12531449946655]
  We propose a Reranking-Truncation joint model (GenRT) that can perform the two tasks concurrently. GenRT integrates reranking and truncation via generative paradigm based on encoder-decoder architecture. Our method achieves SOTA performance on both reranking and truncation tasks for web search and retrieval-augmented LLMs.
  arXiv  Detail & Related papers  (2024-02-05T06:52:53Z)
• Corrective Retrieval Augmented Generation [36.04062963574603]
  Retrieval-augmented generation (RAG) relies heavily on relevance of retrieved documents, raising concerns about how the model behaves if retrieval goes wrong. We propose the Corrective Retrieval Augmented Generation (CRAG) to improve the robustness of generation. CRAG is plug-and-play and can be seamlessly coupled with various RAG-based approaches.
  arXiv  Detail & Related papers  (2024-01-29T04:36:39Z)
• DAPR: A Benchmark on Document-Aware Passage Retrieval [57.45793782107218]
  We propose and name this task emphDocument-Aware Passage Retrieval (DAPR) While analyzing the errors of the State-of-The-Art (SoTA) passage retrievers, we find the major errors (53.5%) are due to missing document context. Our created benchmark enables future research on developing and comparing retrieval systems for the new task.
  arXiv  Detail & Related papers  (2023-05-23T10:39:57Z)
• Natural Logic-guided Autoregressive Multi-hop Document Retrieval for Fact Verification [21.04611844009438]
  We propose a novel retrieve-and-rerank method for multi-hop retrieval. It consists of a retriever that jointly scores documents in the knowledge source and sentences from previously retrieved documents. It is guided by a proof system that dynamically terminates the retrieval process if the evidence is deemed sufficient.
  arXiv  Detail & Related papers  (2022-12-10T11:32:38Z)
• Revisiting DocRED -- Addressing the False Negative Problem in Relation Extraction [39.78594332093083]
  We re-annotate 4,053 documents in the DocRED dataset by adding the missed relation triples back to the original DocRED. We conduct extensive experiments with state-of-the-art neural models on both datasets, and the experimental results show that the models trained and evaluated on our Re-DocRED achieve performance improvements of around 13 F1 points.
  arXiv  Detail & Related papers  (2022-05-25T11:54:48Z)
• Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED [60.39125850987604]
  We show that a textit-revise scheme results in false negative samples and an obvious bias towards popular entities and relations. The relabeled dataset is released to serve as a more reliable test set of document RE models.
  arXiv  Detail & Related papers  (2022-04-17T11:29:01Z)
• GERE: Generative Evidence Retrieval for Fact Verification [57.78768817972026]
  We propose GERE, the first system that retrieves evidences in a generative fashion. The experimental results on the FEVER dataset show that GERE achieves significant improvements over the state-of-the-art baselines.
  arXiv  Detail & Related papers  (2022-04-12T03:49:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.