China

Autoregressive neural language models (LMs) generate a probability distribution over tokens at each time step given a prompt. In this work, we attempt to systematically understand the probability distributions that LMs can produce, showing that some distributions are significantly harder to elicit than others. Specifically, for any target next-token distribution over the vocabulary, we attempt to find a prompt that induces the LM to output a distribution as close as possible to the target, using either soft or hard gradient-based prompt tuning. We find that (1) in general, distributions with very low or very high entropy are easier to approximate than those with moderate entropy; (2) among distributions with the same entropy, those containing &#39;&#39;outlier tokens&#39;&#39; are easier to approximate; (3) target distributions generated by LMs -- even LMs with different tokenizers -- are easier to approximate than randomly chosen targets. These results offer insights into the expressiveness of LMs and the challenges of using them as probability distribution proposers.

EMNLP 2025

Distribution Prompting: Understanding the Expressivity of Language Models Through the Next-Token Distributions They Can Produce

prefix tuning

interpretability

adversarial attacks

## Welcome to EMNLP 2024! 
We are excited to welcome you to one of the most prominent conferences in the field of Natural Language Processing. This year, EMNLP 2024 is being held in a hybrid format,
offering both virtual and in-person participation in beautiful Miami. Due to a record-breaking number of submissions, we've expanded the total number of accepted papers to accommodate more cutting-edge research from around the globe.
### [Conference Handbook](https://drive.google.com/file/d/1WPROgxjLAC96AJL7Ugy0tEnYm7dkrbHt/view?usp=sharing)

You are required to register for this event. **Please register [here](https://2024.emnlp.org/registration/).** The EMNLP 2024 event page on Underline will be open to public one week prior to the event.

Please register!

EMNLP 2024

EMNLP 2024 will take place in Miami, Florida from Nov 12th to Nov 16th, 2024, at the Hyatt Regency Miami Hote and on Underline for remote participants.

Autoregressive neural language models (LMs) generate a probability distribution over tokens at each time step given a prompt. In this work, we attempt to systematically understand the probability distributions that LMs can produce, showing that some distributions are significantly harder to elicit than others. Specifically, for any target next-token distribution over the vocabulary, we attempt to find a prompt that induces the LM to output a distribution as close as possible to the target, using either soft or hard gradient-based prompt tuning. We find that (1) in general, distributions with very low or very high entropy are easier to approximate than those with moderate entropy; (2) among distributions with the same entropy, those containing ''outlier tokens'' are easier to approximate; (3) target distributions generated by LMs -- even LMs with different tokenizers -- are easier to approximate than randomly chosen targets. These results offer insights into the expressiveness of LMs and the challenges of using them as probability distribution proposers.

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.

With the development and widespread application of large language models (LLMs), the new paradigm of ``Model as Product'' is rapidly evolving, and demands higher capabilities to address complex user needs, often requiring precise workflow execution which involves the accurate understanding of multiple tasks. However, existing benchmarks focusing on single-task environments with limited constraints, lack the complexity required to fully reflect To bridge this gap, we present the Extremely Complex Instruction Following Benchmark (EIFBENCH), meticulously crafted to facilitate a more realistic and robust evaluation of LLMs. EIFBENCH not only includes multi-task scenarios that enable comprehensive assessment across diverse task types concurrently, but also integrates a variety of constraints, replicating complex operational environments. Furthermore, we propose the Segment Policy Optimization (SegPO) algorithm to enhance the LLM's ability to accurately fulfill multi-task workflow. Evaluations on EIFBENCH have unveiled considerable performance discrepancies in existing LLMs when challenged with these extremely complex instructions. This finding underscores the necessity for ongoing optimization to navigate the intricate challenges posed by real-world LLM applications.

EIFBENCH: Extremely Complex Instruction Following Benchmark for Large Language Models

Large language models (LLMs) have made significant breakthroughs in extracting useful information from conversation history to enhance the response in long-term conversations. Summarizing useful information from historical conversations has achieved remarkable performance, which, however, may introduce irrelevant or redundant information, making it difficult to flexibly choose and integrate key information from different sessions during memory retrieval. To address this issue, we propose a Fragment-then-Compose framework, a novel memory utilization approach for long-term open-domain conversation, called *FraCom*. To be specific, inspired by the concept of proposition representation from Cognitive Psychology, we first represent the conversation history as a series of predicates plus arguments for propositional representation to preserve key information useful for memory ("**Fragment**"). Then, we compose propositional graphs for the conversation history based on the connection between shared arguments ("**Compose**"). This essentially allows for flexible and effective utilization of related information in long-term memory for better response generation towards a query. Experimental results on four long-term open-domain conversation datasets demonstrate the effectiveness of our *FraCom* in memory utilization and its ability to enhance response generation for LLMs.

Flexibly Utilize Memory for Long-Term Conversation via a Fragment-then-Compose Framework

Evaluating creativity is challenging, even for humans, not only because of its subjectivity but also because it involves complex cognitive processes. Inspired by work in marketing, we attempt to break down creativity into atypicality and originality. With fine-grained human annotations on these dimensions, we propose a suit of tasks specifically for such a subjective problem. We also evaluate the alignment between state-of-the-art (SoTA) vision language models (VLM) and humans on our proposed benchmark, demonstrating both the promises and challenges of using VLMs for automatic creativity assessment.

Leveraging Large Models to Evaluate Novel Content: A Case Study on Advertisement Creativity

As large language models (LLMs) see wider real-world use, understanding and mitigating their unsafe behaviors is critical. Interpretation techniques can reveal causes of unsafe outputs and guide safety, but such connections with safety are often overlooked in prior surveys. We present the first survey that bridges this gap, introducing a unified framework that connects safety-focused interpretation methods, the safety enhancements they inform, and the tools that operationalize them. Our novel taxonomy, organized by LLM workflow stages, summarizes nearly 70 works at their intersections. We conclude with open challenges and future directions. This timely survey helps researchers and practitioners navigate key advancements for safer, more interpretable LLMs.

Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safety

We propose a novel framework, Continuous-Time Attention, which infuses partial differential equations (PDEs) into the Transformer’s attention mechanism to address the challenges of extremely long input sequences. Instead of relying solely on a static attention matrix, we allow attention weights to evolve over a pseudo-time dimension via diffusion, wave, or reaction-diffusion dynamics. This mechanism systematically smooths local noise, enhances long-range dependencies, and stabilizes gradient flow. Theoretically, our analysis shows that PDE-based attention leads to better optimization landscapes and polynomial rather than exponential decay of distant interactions. Empirically, we benchmark our method on diverse experiments—demonstrating consistent gains over both standard and specialized long-sequence Transformer variants. Our findings highlight the potential of PDE-based formulations to enrich attention mechanisms with continuous-time dynamics and global coherence.

Continuous-Time Attention: PDE-Guided Mechanisms for Long-Sequence Transformers

Large language models (LLMs) often appear to excel on public benchmarks, but these high scores may mask an overreliance on dataset-specific surface cues rather than true language understanding. We introduce the Chameleon Benchmark Overfit Detector (C-BOD), a meta-evaluation framework designed to reveal such overfitting. C-BOD systematically rephrases benchmark inputs via a parameterized transformation that preserves semantic content and labels, enabling the detection of performance degradation indicative of superficial pattern reliance. We conduct extensive experiments across two datasets, three rephrasing models, and multiple distortion levels, evaluating 27 state-of-the-art LLMs. On the MMLU benchmark, C-BOD reveals an average performance drop of 2.75% under modest rephrasings, with over 80% of models exhibiting statistically significant differences. Notably, higher-performing models and larger LLMs tend to show greater sensitivity, suggesting a deeper dependence on benchmark-specific phrasing. Due to its dataset and model-agnostic design, C-BOD can be easily integrated into evaluation pipelines and offers a promising foundation for overfitting mitigation strategies. Our findings challenge the community to look beyond leaderboard scores and prioritize resilience and generalization in LLM evaluation.

Forget What You Know about LLMs Evaluations - LLMs are Like a Chameleon

Idioms have long posed a challenge due to their unique linguistic properties, which set them apart from other common expressions. While recent studies have leveraged large language models (LLMs) to handle idioms across various tasks, e.g., idiom-containing sentence generation and idiomatic machine translation, little is known about the underlying mechanisms of idiom processing in LLMs, particularly in multilingual settings. To this end, we introduce MIDAS, a new large-scale dataset of idioms in six languages, each paired with its corresponding meaning. Leveraging this resource, we conduct a comprehensive evaluation of LLMs’ idiom processing ability, identifying key factors that influence their performance. Our findings suggest that LLMs rely not only on memorization, but also adopt a hybrid approach that integrates contextual cues and reasoning, especially when processing compositional idioms. This implies that idiom understanding in LLMs emerges from an interplay between internal knowledge retrieval and reasoning-based inference.

Memorization or Reasoning? Exploring the Idiom Understanding of LLMs

This paper addresses the important yet underexplored task of multi-class sentiment analysis (MCSA), which remains challenging due to subtle semantic differences between adjacent sentiment categories and the scarcity of high-quality annotated data. To tackle these challenges, RD-MCSA (Rationales and Demonstrations-based Multi-Class Sentiment Analysis) is proposed as an In-Context Learning (ICL) framework designed to improve MCSA performance under limited supervision by integrating classification rationales and adaptively selected demonstrations. First, semantically grounded classification rationales are generated from a representative, class-balanced subset of annotated samples selected using a tailored balanced coreset algorithm. These rationales are then paired with demonstrations selected via a similarity-based mechanism powered by a multi-kernel Gaussian process (MK-GP), enabling large language models (LLMs) to better capture fine-grained sentiment distinctions. Experiments on five benchmark sentiment datasets show that RD-MCSA consistently outperforms both supervised baselines and standard ICL methods across various evaluation metrics.

RD-MCSA: A Multi-Class Sentiment Analysis Approach Integrating In-Context Classification Rationales and Demonstrations

We introduce CREPE (Rapid Chest X-ray Report Evaluation by Predicting Multi-category Error Counts), a rapid, interpretable, and clinically grounded metric for automated chest X-ray report generation. CREPE uses a domain-specific BERT model fine-tuned with a multi-head regression architecture to predict error counts across six clinically meaningful categories. Trained on a large-scale synthetic dataset of 32,000 annotated report pairs, CREPE demonstrates strong generalization and interpretability. On the expert-annotated ReXVal dataset, CREPE achieves a Kendall's tau correlation of 0.786 with radiologist error counts, outperforming traditional and recent metrics. CREPE achieves these results with an inference speed approximately 280 times faster than large language model (LLM)-based approaches, enabling rapid and fine-grained evaluation for scalable development of chest X-ray report generation models.

CREPE: Rapid Chest X-ray Report Evaluation by Predicting Multi-category Error Counts

Large language models often retain unintended content, prompting growing interest in knowledge unlearning. Recent approaches emphasize localized unlearning, which restricts parameter updates to specific regions in an effort to remove target knowledge while preserving unrelated general knowledge. However, their effectiveness remains uncertain due to the lack of robust and thorough evaluation of the trade-off between the competing goals of unlearning. In this paper, we begin by revisiting existing localized unlearning approaches. We then conduct controlled experiments to rigorously evaluate whether local parameter updates causally contribute to unlearning. Our findings reveal that the set of parameters that must be modified for effective unlearning is not strictly determined, challenging the core assumption of localized unlearning that parameter locality is inherently indicative of effective knowledge removal.

Downloads

Next from EMNLP 2025

EIFBENCH: Extremely Complex Instruction Following Benchmark for Large Language Models

Similar lecture

What Are the Odds? Language Models Are Capable of Probabilistic Reasoning

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES