China

Despite continuous advancements in the capabilities of large language models (LLMs), numerical reasoning remains a challenging area. Techniques like chain-of-thought prompting, tree-of-thought prompting, and program-of-thought prompting guide LLMs through intermediate reasoning steps. Although in-context learning with few-shot prompting has improved performance, LLMs still lag behind state-of-the-art models on financial numerical reasoning datasets such as FinQA and ConvFinQA. In this work, we introduce FINDER, a novel two-step framework, to enhance LLM&#39;s capabilities in financial numerical reasoning. The first step utilizes a generative retriever to extract relevant facts from unstructured data, including both text and tables. This is followed by context-aware Program of Thought prompting with dynamic selection of in-context examples. Our model FINDER achieves a new state-of-the-art performance on both the FinQA and ConvFinQA datasets, surpassing previous benchmarks with execution accuracy improvements of 5.98% and 4.05%, respectively.

EMNLP 2025

Program of Thoughts for Financial Reasoning: Leveraging Dynamic In-Context Examples and Generative Retrieval

program of thought

fine tuning

**Welcome to EMNLP 2023!**

On behalf of the EMNLP 2023 Organizing Committee, I extend a warm and heartfelt welcome to all of you to the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP). It is with immense pleasure and excitement that we gather here in Singapore, a vibrant hub of innovation and technological advancement.

The conference program is packed with insightful presentations, thought-provoking workshops, and engaging networking opportunities. In addition to the technical sessions, we have also planned several social events that will provide you with the opportunity to connect with your colleagues.

The conference is held in person at the Resorts World Convention Centre in Singapore, and available on-line with the help of Underline.

For those in Singapore, I hope that you will find time to explore the exciting Sentosa Island, Gardens by the Bay, Singapore Botanic Gardens and the many other attractions unique to Singapore.

Of course, we are grateful to our sponsors and partners for their generous support of EMNLP 2023, whose contributions make it possible for us to host this world-class event.

Yuji Matsumoto (RIKEN AIP) 
EMNLP 2023 General Chair

To access the **EMNLP 2023** event page on Underline, you need to register for the Conference. 
Please follow **[this link](https://2023.emnlp.org/registration/)** for more details.

EMNLP 2023

EMNLP 2023 took place in Singapore from Dec 6th to Dec 10th, 2023.

Despite continuous advancements in the capabilities of large language models (LLMs), numerical reasoning remains a challenging area. Techniques like chain-of-thought prompting, tree-of-thought prompting, and program-of-thought prompting guide LLMs through intermediate reasoning steps. Although in-context learning with few-shot prompting has improved performance, LLMs still lag behind state-of-the-art models on financial numerical reasoning datasets such as FinQA and ConvFinQA. In this work, we introduce FINDER, a novel two-step framework, to enhance LLM's capabilities in financial numerical reasoning. The first step utilizes a generative retriever to extract relevant facts from unstructured data, including both text and tables. This is followed by context-aware Program of Thought prompting with dynamic selection of in-context examples. Our model FINDER achieves a new state-of-the-art performance on both the FinQA and ConvFinQA datasets, surpassing previous benchmarks with execution accuracy improvements of 5.98% and 4.05%, respectively.

poster

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community." 

*Message from the General Chair, Dirk Hovy*

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

Dense Passage Retrieval (DPR) typically relies on Euclidean or Cosine distance to measure query-passage relevance in embedding space. While effective when embeddings lie on a linear manifold, our experiments across DPR benchmarks suggest that embeddings often lie on lower-dimensional, non-linear manifolds, especially in out-of-distribution (OOD) settings, where these distances fail to capture semantic similarity. To address this limitation, we propose a manifold-aware distance metric for DPR (MA-DPR) that models the intrinsic manifold structure of passages using a nearest neighbor graph and measures distance between query and passages based on their shortest path in this graph. We show that MA-DPR outperforms Euclidean and Cosine distance by up to 26% on OOD passage retrieval while maintaining performance on in-distribution data across various embedding models, with only a small increase in query inference time. Empirical evidence suggests that manifold-aware distance allows DPR to leverage context from related neighboring passages, making it effective even in the absence of direct semantic overlap. In addition, it can be extended to a wide range of dense embedding and DPR tasks, offering practical utility across diverse retrieval scenarios.

MA-DPR: Manifold-aware Distance Metrics for Dense Passage Retrieval

Temporal relation extraction (TRE) is a fundamental task in natural language processing (NLP) that involves identifying the temporal relationships between events in a document. Despite the advances in large language models (LLMs), their application to TRE remains limited. Most existing approaches rely on pairwise classification, where event pairs are classified in isolation, leading to computational inefficiency and a lack of global consistency in the resulting temporal graph. In this work, we propose a novel zero-shot method for TRE that generates a document’s complete temporal graph in a single step, followed by temporal constraint optimization to refine predictions and enforce temporal consistency across relations. Additionally, we introduce OmniTemp, a new dataset with complete annotations for all pairs of targeted events within a document. Through experiments and analyses, we demonstrate that our method outperforms existing zero-shot approaches and offers a competitive alternative to supervised TRE models.

Beyond Pairwise: Global Zero-shot Temporal Graph Generation

In pharmacovigilance, effective automation of medical data structuring, especially linking entities to standardized terminologies such as MedDRA, is critical. This challenge is rarely addressed for German data. With MedLinkDE we address German MedDRA entity linking for adverse drug reactions in a two-step approach: (1) retrieval of medical terms with fine-tuned embedding models, followed (2) by guided chain-of-thought re-ranking using LLMs. To this end, we introduce RENOde, a German real-world MedDRA dataset consisting of reportings from patients and healthcare professionals. To overcome the challenges posed by the linguistic diversity of these reports, we generate synthetic data mapping the two reporting styles of patients and healthcare professionals. Our embedding models, fine-tuned on these synthetic, quasi-personalized datasets, show competitive performance with real datasets in terms of accuracy at high top- recall, providing a robust basis for re-ranking. Our subsequent guided Chain of Thought (CoT) re-ranking, informed by MedDRA coding guidelines, improves entity linking accuracy by approximately 15% (Acc@1) compared to embedding-only strategies. In this way, our approach demonstrates the feasibility of entity linking in medical reports under the constraints of data scarcity by relying on synthetic data reflecting different informant roles of reporting persons.

MedLinkDE – MedDRA Entity Linking for German with Guided Chain of Thought Reasoning

The integration of external knowledge through Retrieval-Augmented Generation (RAG) has become foundational in enhancing large language models (LLMs) for knowledge-intensive tasks. However, existing RAG paradigms often overlook the cognitive step of applying knowledge, leaving a gap between retrieved facts and task-specific reasoning. In this work, we introduce RAG+, a principled and modular extension that explicitly incorporates application-aware reasoning into the RAG pipeline. RAG+ constructs a dual corpus consisting of knowledge and aligned application examples, created either manually or automatically, and jointly retrieves both during inference. This design enables LLMs not only to access relevant information but also to apply it within structured, goal-oriented reasoning processes. Experiments across mathematical, law, and medical domains, conducted on multiple models, demonstrate that RAG+ consistently outperforms standard RAG variants, achieving average improvements of 3–5%, and peak gains up to 7.5% in complex scenarios. By bridging retrieval with actionable application, RAG+ advances a more cognitively grounded framework for knowledge integration, representing a step toward more interpretable and capable LLMs.

RAG+: Enhancing Retrieval-Augmented Generation with Application-Aware Reasoning

The widespread dissemination of toxic content on social media poses a serious threat to both online environments and public discourse, highlighting the urgent need for detoxification methods that effectively remove toxicity while preserving the original semantics. However, existing approaches often struggle to simultaneously achieve strong detoxification performance, semantic preservation, and robustness to out-of-distribution data. Moreover, they typically rely on costly, manually annotated parallel corpora while showing poor data efficiency. To address these challenges, we propose a two-stage training framework that jointly optimizes for Data Efficiency, Semantic Preservation, and Model Generalization. We first perform supervised fine-tuning on a small set of high-quality, filtered parallel data to establish a strong initialization. Then, we leverage unlabeled toxic inputs and a custom-designed reward model to train the LLM using Group Relative Policy Optimization. Experimental results demonstrate that our method effectively mitigates the trade-offs faced by previous work, achieving state-of-the-art performance with improved generalization and significantly reduced dependence on annotated data. Our code is available at https://github.com/allacnobug/Detoxification-of-Text.

Text Detoxification: Data Efficiency, Semantic Preservation and Model Generalization

Do LLMs genuinely incorporate external definitions, or do they primarily rely on their parametric knowledge? To address these questions, we conduct controlled experiments across multiple explanation benchmark datasets (general and domain-specific) and label definition conditions, including expert-curated, LLM-generated, perturbed, and swapped definitions. Our results reveal that while explicit label definitions can enhance accuracy and explainability, their integration into an LLM’s task-solving processes is neither guaranteed nor consistent, suggesting reliance on internalized representations in many cases. Models often default to their internal representations, particularly in general tasks, whereas domain-specific tasks benefit more from explicit definitions. These findings underscore the need for a deeper understanding of how LLMs process external knowledge alongside their pre-existing capabilities.

Do LLMs Adhere to Label Definitions? Examining Their Receptivity to External Label Definitions

With the rise of long-context language models (LMs) capable of processing tens of thousands of tokens in a single context window, do multi-stage retrieval-augmented generation (RAG) pipelines still offer measurable benefits over simpler, single-stage approaches? To assess this question, we conduct a controlled evaluation for QA tasks under systematically scaled token budgets, comparing two recent multi-stage pipelines, ReadAgent and RAPTOR, against three baselines, including DOS RAG (Document's Original Structure RAG), a simple retrieve-then-read method that preserves original passage order. Despite its straightforward design, DOS RAG consistently matches or outperforms more intricate methods on multiple long-context QA benchmarks. We trace this strength to a combination of maintaining source fidelity and document structure, prioritizing recall within effective context windows, and favoring simplicity over added pipeline complexity. We recommend establishing DOS RAG as a simple yet strong baseline for future RAG evaluations, paired with state-of-the-art embedding and language models, and benchmarked under matched token budgets, to ensure that added pipeline complexity is justified by clear performance gains as models continue to improve.

Stronger Baselines for Retrieval-Augmented Generation with Long-Context Language Models

Given some information need, Large Language Models (LLMs) are increasingly used for candidate text relevance assessment, typically using a one-by-one pointwise (PW) strategy where each LLM call evaluates one candidate at a time. Meanwhile, it has been shown that LLM performance can be improved through self-consistency: prompting the LLM to do the same task multiple times (possibly in perturbed ways) and then aggregating the responses. To take advantage of self-consistency, we hypothesize that batched PW strategies, where multiple passages are judged in one LLM call, are better suited than one-by-one PW methods since a larger input context can induce more diverse LLM sampling across self-consistency calls. We first propose several candidate batching strategies to create prompt diversity across self-consistency calls through subset reselection and permutation. We then test our batched PW methods on relevance assessment and ranking tasks against one-by-one PW and listwise LLM ranking baselines with and without self-consistency, using three passage retrieval datasets and GPT-4o, Claude Sonnet 3, and Amazon Nova Pro. We find that batched PW methods outperform all baselines, and show that batching can greatly amplify the positive effects of self-consistency. For instance, on our legal search dataset, GPT-4o one-by-one PW ranking NDCG@10 improves only from 44.9% to 46.8% without self-consistency vs. with 15 self consistency calls, while batched PW ranking improves from 43.8% to 51.3%, respectively.

Batched Self-Consistency Improves LLM Relevance Assessment and Ranking

We introduce SemCSE, an unsupervised method for learning semantic embeddings of scientific texts. Building on recent advances in contrastive learning for text embeddings, our approach leverages LLM-generated summaries of scientific abstracts to train a model that positions semantically related summaries closer together in the embedding space. This resulting objective ensures that the model captures the true semantic content of a text, in contrast to traditional citation-based approaches that do not necessarily reflect semantic similarity. To validate this, we propose a novel benchmark designed to assess a model's ability to understand and encode the semantic content of scientific texts, demonstrating that our method enforces a stronger semantic separation within the embedding space. Additionally, we evaluate SemCSE on the comprehensive SciRepEval benchmark for scientific text embeddings, where it achieves state-of-the-art performance among models of its size, thus highlighting the benefits of a semantically focused training approach.

SemCSE: Semantic Contrastive Sentence Embeddings Using LLM-Generated Summaries For Scientific Abstracts

In exploratory search, users often submit vague queries to investigate unfamiliar topics, but receive limited feedback about how the search engine understood their input. This leads to a self-reinforcing cycle of mismatched results and trial-and-error reformulation. To address this, we study the task of generating user-facing natural language query intent descriptions that surface what the system likely inferred the query to mean, based on post-retrieval evidence. We propose QUIDS, a method that leverages dual-space contrastive learning to isolate intent-relevant information while suppressing irrelevant content. QUIDS combines a dual-encoder representation space with a disentangling decoder that works together to produce concise and accurate intent descriptions. Enhanced by intent-driven hard negative sampling, the model significantly outperforms state-of-the-art baselines across ROUGE, BERTScore, and human/LLM evaluations. Our qualitative analysis confirms QUIDS' effectiveness in generating accurate intent descriptions for exploratory search. Our work contributes to improving the interaction between users and search engines by providing feedback to the user in exploratory search settings.\footnote{Our code is available at \url{https://anonymous.4open.science/r/QID-6E00/}}

Downloads

Next from EMNLP 2025

MA-DPR: Manifold-aware Distance Metrics for Dense Passage Retrieval

Similar lecture

What Makes Chain-of-Thought Prompting Effective? A Counterfactual Study