China

Large Language Models (LLMs) have emerged as powerful sources of evidence for linguists seeking to develop theories of syntax. In this paper, we argue that causal interpretability methods, applied to LLMs, can greatly enhance the value of such evidence by helping us characterize the abstract mechanisms that LLMs learn to use. Our empirical focus is a set of English filler--gap dependency constructions (e.g., questions, relative clauses). Linguistic theories largely agree that these constructions share many properties. Using experiments based in Distributed Interchange Interventions, we show that LLMs converge on similar abstract analyses of these constructions. These analyses also reveal previously overlooked factors -- relating to frequency, filler type, and surrounding context -- that could motivate changes to standard linguistic theory. Overall, these results suggest that mechanistic, internal analyses of LLMs can push linguistic theory forward.

EMNLP 2025

Causal Interventions Reveal Shared Structure Across English Filler--Gap Constructions

computational psycholinguistics

linguistic theories

cognitive modeling

## 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."<br>

*Message from the General Chair, Dirk Hovy*





<br>

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

technical paper

Recent advances in text-to-speech technologies have enabled realistic voice generation, fueling audio-based deepfake attacks such as fraud and impersonation. While audio anti-spoofing systems are critical for detecting such threats, prior work has predominantly focused on acoustic-level perturbations, leaving the **impact of linguistic variation largely unexplored**. In this paper, we investigate the linguistic sensitivity of both open-source and commercial anti-spoofing detectors by introducing transcript-level adversarial attacks. Our extensive evaluation reveals that even minor linguistic perturbations can significantly degrade detection accuracy: attack success rates surpass **60%** on several open-source detector–voice pairs, and notably one commercial detection accuracy drops from **100%** on synthetic audio to just **32%**. Through a comprehensive feature attribution analysis, we identify that both linguistic complexity and model-level audio embedding similarity contribute strongly to detector vulnerability. We further demonstrate the real-world risk via a case study replicating the Brad Pitt audio deepfake scam, using transcript adversarial attacks to completely bypass commercial detectors. These results highlight the need to **move beyond purely acoustic defenses and account** for linguistic variation in the design of robust anti-spoofing systems. All source code will be publicly available.

What You Read Isn't What You Hear: Linguistic Sensitivity in Deepfake Speech Detection

Humans have the ability to utilize visual cues, such as lip movements and visual scenes, to enhance auditory perception, particularly in noisy environments. However, current Automatic Speech Recognition (ASR) or Audio-Visual Speech Recognition (AVSR) models often struggle in noisy scenarios. To solve this task, we propose a model that improves transcription by correlating noise sources to visual cues. Unlike works that rely on lip motion and require the speaker's visibility, we exploit broader visual information from the environment. This allows our model to naturally filter speech from noise and improve transcription, much like humans do in noisy scenarios. Our method re-purposes pretrained speech and visual encoders, linking them with multi-headed attention. This approach enables the transcription of speech and the prediction of noise labels in video inputs. We introduce a scalable pipeline to develop audio-visual datasets, where visual cues correlate to noise in the audio. We show significant improvements over existing audio-only models in noisy scenarios. Results also highlight that visual cues play a vital role in improved transcription accuracy

Visual-Aware Speech Recognition for Noisy Scenarios

Large Language Models (LLMs) have been shown to achieve breakthrough performances on complex logical reasoning tasks. Nevertheless, most existing research focuses on employing formal language to guide LLMs for deriving reliable reasoning paths, with systematic evaluations of these capabilities still being limited. In this paper, we aim to conduct a comprehensive evaluation of LLMs across various logical reasoning problems utilizing formal languages. From the perspective of three dimensions, i.e., spectrum of LLMs, taxonomy of tasks, and format of trajectories, our key findings are: 1. \textit{Thinking} models significantly outperform \textit{Instruct} models, especially when formal language is employed; 2. All LLMs exhibit limitations in inductive reasoning capability, irrespective of whether they use a formal language; 3. Data with PoT format achieves the best generalization performance across other languages. Additionally, we also curate the formal-relative training data to further enhance the small language models, and the experimental results indicate that a simple rejected fine-tuning method can better enable LLMs to generalize across formal languages and achieve the best overall performance.

Do Large Language Models excel in Complex Logical Reasoning with Formal Language?

Evaluating statement autoformalization, translating natural language mathematics into formal languages like Lean 4, remains a significant challenge, with few metrics, datasets, and standards to robustly measure progress. In this work, we present a comprehensive approach combining improved metrics, robust benchmarks, and systematic evaluation, to fill this gap. First, we introduce BEq+, an automated metric that correlates strongly with human judgment, along with ProofNetVerif, a new dataset for assessing the quality of evaluation metrics, containing 3,752 annotated examples. Second, we develop two new autoformalization benchmarks: ProofNet#, a corrected version of ProofNet, and RLM25, with 619 new pairs of research-level mathematics from six formalization projects. Through systematic experimentation across these benchmarks, we find that current techniques can achieve up to 45.1% accuracy on undergraduate mathematics but struggle with research-level content without proper context. Our work establishes a reliable foundation for evaluating and advancing autoformalization systems.

Reliable Evaluation and Benchmarks for Statement Autoformalization

Large Language Models (LLMs) have shown growing potential in molecular sciences, but they often produce chemically inaccurate descriptions and struggle to recognize or justify potential errors. This raises important concerns about their robustness and reliability in scientific applications. To support more rigorous evaluation of LLMs in chemical reasoning, we present the MolErr2Fix benchmark, designed to assess LLMs on error detection and correction in molecular descriptions. Unlike existing benchmarks focused on molecule-to-text generation or property prediction, MolErr2Fix emphasizes fine-grained chemical understanding. It tasks LLMs with identifying, localizing, explaining, and revising potential structural and semantic errors in molecular descriptions. Specifically, MolErr2Fix consists of 1,193 fine-grained annotated error instances. Each instance contains quadruple annotations, i.e., (error type, span location, the explanation, and the correction). These tasks are intended to reflect the types of reasoning and verification required in real-world chemical communication. Evaluations of current state-of-the-art LLMs reveal notable performance gaps, underscoring the need for more robust chemical reasoning capabilities. MolErr2Fix provides a focused benchmark for evaluating such capabilities and aims to support progress toward more reliable and chemically informed language models. All annotations and an accompanying evaluation API will be publicly released to facilitate future research.

MolErr2Fix: Benchmarking LLM Trustworthiness in Chemistry via Modular Error Detection, Localization, Explanation, and Correction

Retrieval-augmented generation (RAG) and long-context language models (LCLMs) both address context limitations of LLMs in open-domain QA. However, optimal external context to retrieve remains an open problem: fixed retrieval budgets risk wasting tokens or omitting key evidence. Existing adaptive methods like Self-RAG and Self-Route rely on iterative LLM prompting and perform well on factoid QA, but struggle with aggregation QA where optimal context size is unknown and variable. We present Adaptive‑k retrieval, a simple and effective single-pass method that selects a query-specific number of passages by applying a threshold to the similarity scores between the query and candidate passages. It does not require model fine-tuning, extra LLM calls or changes to existing retriever–reader pipelines. On both factoid and aggregation QA benchmarks, Adaptive‑k matches or outperforms fixed‑k baselines while using up to 10x fewer tokens than full-context input, and still retrieves 70% of relevant passages. It improves accuracy across five LCLMs and two embedding models, highlighting that dynamically adjusting context size leads to more efficient and accurate QA.

Efficient Context Selection for Long-Context QA: No Tuning, No Iteration, Just Adaptive‑k

Large language models (LLMs) exhibit prompt leakage vulnerabilities, where they may be coaxed into revealing system prompts embedded in LLM services, raising intellectual property and confidentiality concerns. An intriguing question arises: Do LLMs genuinely internalize prompt leakage intents in their hidden states before generating tokens? In this work, we use probing techniques to capture LLMs' intent-related internal representations and confirm that the answer is yes. We start by comprehensively inducing prompt leakage behaviors across diverse system prompts, attack queries, and decoding methods. We develop a hybrid labeling pipeline, enabling the identification of broader prompt leakage behaviors beyond mere verbatim leaks. Our results show that a simple linear probe can predict prompt leakage risks from pre-generation hidden states without generating any tokens. Across all tested models, linear probes consistently achieve 90%+ AUROC, even when applied to new system prompts and attacks. Understanding the model internals behind prompt leakage drives practical applications, including intention-based detection of prompt leakage risks. Code is available at: https://anonymous.4open.science/r/Probing-leakage-intents.

“I’ve Decided to Leak”: Probing Internals Behind Prompt Leakage Intents

Generalizing to unseen graph tasks without task-specific supervision remains challenging. Graph Neural Networks (GNNs) are limited by fixed label spaces, while Large Language Models (LLMs) lack structural inductive biases. Recent advances in Large Reasoning Models (LRMs) provide a zero-shot alternative via explicit, long chain-of-thought reasoning. Inspired by this, we propose a GNN-free approach that reformulates graph tasks—node classification, link prediction, and graph classification—as textual reasoning problems solved by LRMs. We introduce the first datasets with detailed reasoning traces for these tasks and develop Graph-R1, a reinforcement learning framework that leverages task-specific rethink templates to guide reasoning over linearized graphs. Experiments demonstrate that Graph-R1 outperforms state-of-the-art baselines in zero-shot settings, producing interpretable and effective predictions. Our work highlights the promise of explicit reasoning for graph learning and provides new resources for future research. Codes are available at https://anonymous.4open.science/r/emnlp_submission-FDF0.

Graph-R1: Incentivizing the Zero-Shot Graph Learning Capability in LLMs via Explicit Reasoning

In natural languages, frequency and compositionality exhibit an inverse relationship: the most frequent words often resist regular patterns, developing idiosyncratic forms. This phenomenon, exemplified by irregular verbs, raises a compelling question: do artificial communication systems follow similar principles? Through systematic experiments with neural network agents in a referential game setting, and by manipulating input frequency through Zipfian distributions, we investigate if these systems mirror the irregular verbs phenomenon, where messages referring to frequent objects develop less compositional structure than messages referring to rare ones. We establish that compositionality is not an inherent property of the frequency itself and provide compelling evidence that limited data exposure, facilitated through frequency distributions, serves as a fundamental driver for the emergence of compositional structure in communication systems, offering insights into the cognitive and computational pressures that shape linguistic systems.

Frequency & Compositionality in Emergent Communication

Large Language Models (LLMs) are intended to reflect human linguistic competencies. But humans have access to a broad and embodied context, which is key in detecting and resolving linguistic ambiguities, even in isolated text spans. A foundational case of semantic ambiguity is found in the task coreference resolution: how is a pronoun related to an earlier person mention? This skill is implicit in every downstream task, and the presence of ambiguity at this level can alter performance significantly. We show that LLMs can achieve good performance with minimal prompting in both coreference disambiguation and the detection of ambiguity in coreference, however, they cannot do both at the same time. We present the Correct-Detect trade-off: though models have both capabilities and deploy them implicitly, successful performance balancing these two abilities remains elusive.

Downloads

Next from EMNLP 2025

What You Read Isn't What You Hear: Linguistic Sensitivity in Deepfake Speech Detection

Similar lecture

Mind the Gap: How BabyLMs Learn Filler-Gap Dependencies