# Beyond Single-Negative Preference: Multi-Negative DPO for LLM-Centric Historical Entity Linking

arXiv: https://arxiv.org/abs/2609.07379 · HTML: https://mlsanity.com/paper/2609.07379/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 13 of its 34 references that are in the map.
- First posted on arXiv 7 Sep 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 13 of 13)

- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [Autoregressive Entity Retrieval](https://mlsanity.com/paper/2010.00904/) (2020)
- [LiPO: Listwise Preference Optimization through Learning-to-Rank](https://mlsanity.com/paper/2402.01878/) (2024)
- [Scalable Zero-shot Entity Linking with Dense Entity Retrieval](https://mlsanity.com/paper/1911.03814/) (2019)
- [Efficient One-Pass End-to-End Entity Linking for Questions](https://mlsanity.com/paper/2010.02413/) (2020)
- [ReFinED: An Efficient Zero-shot-capable Approach to End-to-End Entity Linking](https://mlsanity.com/paper/2207.04108/) (2022)
- [LLMAEL: Large Language Models are Good Context Augmenters for Entity Linking](https://mlsanity.com/paper/2407.04020/) (2024)
- [Instructed Language Models with Retrievers Are Powerful Entity Linkers](https://mlsanity.com/paper/2311.03250/) (2023)
- [Multilingual End to End Entity Linking](https://mlsanity.com/paper/2306.08896/) (2023)
- [MPPO: Multi Pair-wise Preference Optimization for LLMs with Arbitrary Negative Samples](https://mlsanity.com/paper/2412.15244/) (2024)
- [Leveraging the Power of Large Language Models in Entity Linking via Adaptive Routing and Targeted Reasoning](https://mlsanity.com/paper/2510.20098/) (2025)
- [Evaluating end-to-end entity linking on domain-specific knowledge bases: Learning about ancient technologies from museum collections](https://mlsanity.com/paper/2305.14588/) (2023)

## Abstract (arXiv)

Large language models (LLMs) have recently shown promise for historical entity linking, but preference optimization for this task is often formulated with only one negative candidate per training instance. This discards information from the remaining candidates retrieved for the same mention. We introduce multi-negative direct preference optimisation (MDPO), a reference-based pairwise objective that compares the correct entity with all valid rejected candidates associated with each mention. MDPO preserves the Bradley-Terry formulation of DPO while exploiting the complete candidate set through masked, length-normalised sequence scores. We evaluate MDPO on hipe-2020 and newseye, covering French, German, English, Swedish, and Finnish historical newspaper text. Experiments show that MDPO improves over supervised fine-tuning and single-negative DPO, with particularly strong gains for NIL mentions, semantic ambiguity, OCR noise, and historically difficult names. Further analyses disentangle candidate-generation and selection errors, showing that candidate retrieval remains a key bottleneck for end-to-end entity linking. These results demonstrate that incorporating all within-instance negative candidates is a simple and effective improvement for LLM-based historical entity linking.