Training Dense Retrievers with Multiple Positive Passages

• URL: http://arxiv.org/abs/2602.12727v1
• Date: Fri, 13 Feb 2026 08:56:52 GMT
• Title: Training Dense Retrievers with Multiple Positive Passages
• Authors: Benben Wang, Minghao Tang, Hengran Zhang, Jiafeng Guo, Keping Bi,
• Abstract summary: We present a systematic study of multi-positive optimization objectives for retriever training.<n>We unify representative objectives, including Joint Likelihood (JointLH), Summed Marginal Likelihood (SumMargLH), and Log-Sum-Exp Pairwise (LSEPair) loss.<n>Our results show that LSEPair consistently achieves superior robustness and performance across settings.
• Score: 36.92346009805258
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Modern knowledge-intensive systems, such as retrieval-augmented generation (RAG), rely on effective retrievers to establish the performance ceiling for downstream modules. However, retriever training has been bottlenecked by sparse, single-positive annotations, which lead to false-negative noise and suboptimal supervision. While the advent of large language models (LLMs) makes it feasible to collect comprehensive multi-positive relevance labels at scale, the optimal strategy for incorporating these dense signals into training remains poorly understood. In this paper, we present a systematic study of multi-positive optimization objectives for retriever training. We unify representative objectives, including Joint Likelihood (JointLH), Summed Marginal Likelihood (SumMargLH), and Log-Sum-Exp Pairwise (LSEPair) loss, under a shared contrastive learning framework. Our theoretical analysis characterizes their distinct gradient behaviors, revealing how each allocates probability mass across positive document sets. Empirically, we conduct extensive evaluations on Natural Questions, MS MARCO, and the BEIR benchmark across two realistic regimes: homogeneous LLM-annotated data and heterogeneous mixtures of human and LLM labels. Our results show that LSEPair consistently achieves superior robustness and performance across settings, while JointLH and SumMargLH exhibit high sensitivity to the quality of positives. Furthermore, we find that the simple strategy of random sampling (Rand1LH) serves as a reliable baseline. By aligning theoretical insights with empirical findings, we provide practical design principles for leveraging dense, LLM-augmented supervision to enhance retriever effectiveness.

Related papers

• Data-Centric Interpretability for LLM-based Multi-Agent Reinforcement Learning [39.84288631342219]
  We analyze large-scale reinforcement learning training runs from the sophisticated environment of Full-Press Diplomacy.<n>We introduce Meta-Autointerp, a method for grouping SAE features into interpretable hypotheses about training dynamics.<n>We find that even subjectively interesting and seemingly helpful SAE features may be worse than useless to humans.
  arXiv  Detail & Related papers  (2026-02-05T01:21:22Z)
• Discovering Process-Outcome Credit in Multi-Step LLM Reasoning [3.584086358722852]
  Reinforcement Learning (RL) serves as a potent paradigm for enhancing reasoning capabilities in Large Language Models (LLMs)<n>We propose a novel framework designed to provide continuous reward signals.<n>Our model exhibits superior out-of-distribution robustness, demonstrating promising zero-shot transfer capabilities to unseen and challenging reasoning tasks.
  arXiv  Detail & Related papers  (2026-02-01T05:44:09Z)
• U-MARVEL: Unveiling Key Factors for Universal Multimodal Retrieval via Embedding Learning with MLLMs [24.551034147718312]
  Universal multimodal retrieval (UMR) aims to address complex retrieval tasks where both queries and candidates span diverse modalities.<n>We present a study aimed at uncovering the key factors that drive effective embedding learning for UMR using MLLMs.<n>We introduce a unified framework termed U-MARVEL, which outperforms state-of-the-art competitors on the M-B benchmark.
  arXiv  Detail & Related papers  (2025-07-20T10:27:34Z)
• Omni-Thinker: Scaling Multi-Task RL in LLMs with Hybrid Reward and Task Scheduling [66.0871543682453]
  We present Omni-Thinker, a unified reinforcement learning framework that scales large language models across diverse tasks.<n>Our scheduler orders tasks according to accuracy backward transfer (BWT), reducing forgetting and improving multi-task performance.
  arXiv  Detail & Related papers  (2025-07-20T01:50:16Z)
• The Hidden Link Between RLHF and Contrastive Learning [56.45346439723488]
  We show that Reinforcement Learning from Human Feedback (RLHF) and the simple Direct Preference Optimization (DPO) can be interpreted from the perspective of mutual information (MI)<n>Within this framework, both RLHF and DPO can be interpreted as methods that performing contrastive learning based on the positive and negative samples derived from base model.<n>We propose the Mutual Information Optimization (MIO) to mitigate the late-stage decline in chosen-likelihood observed in DPO.
  arXiv  Detail & Related papers  (2025-06-27T18:51:25Z)
• LTRR: Learning To Rank Retrievers for LLMs [53.285436927963865]
  We show that routing-based RAG systems can outperform the best single-retriever-based systems.<n>Performance gains are especially pronounced in models trained with the Answer Correctness (AC) metric.<n>As part of the SIGIR 2025 LiveRAG challenge, our submitted system demonstrated the practical viability of our approach.
  arXiv  Detail & Related papers  (2025-06-16T17:53:18Z)
• Direct Retrieval-augmented Optimization: Synergizing Knowledge Selection and Language Models [83.8639566087953]
  We propose a direct retrieval-augmented optimization framework, named DRO, that enables end-to-end training of two key components.<n>DRO alternates between two phases: (i) document permutation estimation and (ii) re-weighted, progressively improving RAG components.<n>Our theoretical analysis reveals that DRO is analogous to policy-gradient methods in reinforcement learning.
  arXiv  Detail & Related papers  (2025-05-05T23:54:53Z)
• Fine-Grained Guidance for Retrievers: Leveraging LLMs' Feedback in Retrieval-Augmented Generation [20.420575358183687]
  Retrieval-Augmented Generation (RAG) has proven to be an effective method for mitigating hallucination issues inherent in large language models (LLMs) Previous approaches typically train retrievers based on semantic similarity, lacking optimization for RAG. We propose a novel framework, FiGRet, which leverages the language capabilities of LLMs to construct examples from a more granular, information-centric perspective.
  arXiv  Detail & Related papers  (2024-11-06T14:42:39Z)
• Preference-Based Multi-Agent Reinforcement Learning: Data Coverage and Algorithmic Techniques [65.55451717632317]
  We study Preference-Based Multi-Agent Reinforcement Learning (PbMARL)<n>We identify the Nash equilibrium from a preference-only offline dataset in general-sum games.<n>Our findings underscore the multifaceted approach required for PbMARL.
  arXiv  Detail & Related papers  (2024-09-01T13:14:41Z)
• Uncertainty Aware Learning for Language Model Alignment [97.36361196793929]
  We propose uncertainty-aware learning (UAL) to improve the model alignment of different task scenarios. We implement UAL in a simple fashion -- adaptively setting the label smoothing value of training according to the uncertainty of individual samples. Experiments on widely used benchmarks demonstrate that our UAL significantly and consistently outperforms standard supervised fine-tuning.
  arXiv  Detail & Related papers  (2024-06-07T11:37:45Z)
• Why So Pessimistic? Estimating Uncertainties for Offline RL through Ensembles, and Why Their Independence Matters [35.17151863463472]
  We take a renewed look at how ensembles of $Q$-functions can be leveraged as the primary source of pessimism for offline reinforcement learning (RL) We propose MSG, a practical offline RL algorithm that trains an ensemble of $Q$-functions with independently computed targets based on completely separate networks. Our experiments on the popular D4RL and RL Unplugged offline RL benchmarks demonstrate that MSG with deep ensembles surpasses highly well-tuned state-of-the-art methods by a wide margin.
  arXiv  Detail & Related papers  (2022-05-27T01:30:12Z)

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.