United States

Recent works have analyzed the impact of individual components of neural networks on gendered predictions, often with a focus on mitigating gender bias. However, mechanistic interpretations of gender tend to (i) focus on a very specific gender-related task, such as gendered pronoun prediction, or (ii) fail to distinguish between the production of \textit{factually gendered outputs} (the correct assumption of gender given a word that carries gender as a semantic property) and \textit{gender biased outputs} (based on a stereotype). To address these issues, we curate Gknow, a benchmark to assess gender knowledge and gender bias in language models across different types of gender-related predictions. Gknow allows us to identify and analyze circuits and individual neurons responsible for gendered predictions. We test the impact of neuron ablation on benchmarks for disentangling stereotypical and factual gender (DiFair and the test set of Gknow), as well as StereoSet. Results show that gender bias and factual gender are severely entangled on the level of both circuits and neurons, entailing that ablation is an unreliable debiasing method. Furthermore, we show that benchmarks for evaluating gender bias can hide the decrease in factual gender knowledge that accompanies neuron ablation. We curate Gknow as a contribution to the continuous development of robust gender bias benchmarks.

ACL 2026

GKnow: Measuring the Entanglement of Gender Bias and Factual Gender

## Welcome to NAACL 2022!
The Annual Meeting of the North American Chapter of the Association for Computational Linguistics (NAACL) is taking place July 10-15, 2022 as a hybrid event, in Seattle, WA and online. We are happy to welcome all of you to conference! 
 
The main conference program features oral presentations, in-person and virtual posters and demo sessions, a plenary session for our best paper presentations and awards, keynote presentations, a plenary panel on the place of linguistics and symbolic structures in NLP, and an Industrial Track panel on Careers in NLP.
 
Posters (including Findings of NAACL 2022) and demos are grouped by areas for both the in-person and the virtual sessions. For the virtual component, the talks will be on Zoom and the posters and the demos will be in GatherTown. The Student Research Workshop and will have an oral session and a poster session.
 
The program also features 6 Tutorials and 26 Workshops.
 
 We wish you a wonderful conference!
 
[The NAACL 2022 Organizing Committee](https://2022.naacl.org/committees/organization/)

NAACL 2022

2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics

technical paper

It is my great pleasure and honor to welcome you to the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), held in San Diego, California, USA, July 2–7, 2026. ACL 2026 continues our fields tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.

It is exciting to see the many developments in the field reflected at our conference, such as the advancement of large language models (born out of our community) into many application directions. This is not without challenges, partly due to unsolved problems, and partly due to misuse of the technology. All this is reflected well at ACL 2026. The conference also reflects that the field is not only about advancing technology, but also the science of fundamentally understanding language and how it is use by many different speaker communities. 
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) 
<html><button style="display: inline-flex; align-items: center; justify-content: center; white-space: nowrap; border-radius: 9999px; font-weight: bold; background: #7c3aed; color: white; font-family: 'Space Grotesk', sans-serif; height: 40px; font-size: 16px; padding: 0 20px; border: none; cursor: pointer" onclick="window.open('https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing')">Conference Handbook</button></html>

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://2026.aclweb.org/registration/) first.

Registration Required

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) will take place in San Diego, California, from July 2 - 7, 2026.

We identify semantically coherent, context-consistent network components in large language models (LLMs) using coactivation of sparse autoencoder (SAE) features collected from just a handful of prompts. Focusing on concept-relation prediction tasks, we show that ablating these components for concepts (e.g., countries and words) and relations (e.g., capital city and translation language) changes model outputs in predictable ways, while amplifying these components induces counterfactual responses. Notably, composing relation and concept components yields compound counterfactual outputs. Further analysis reveals that while most concept components emerge from the very first layer, more abstract relation components are concentrated in later layers. Lastly, we show that extracted components more comprehensively capture concepts and relations than individual features while maintaining specificity. Overall, our findings suggest a modular organization of knowledge and advance methods for efficient, targeted LLM manipulation.

Sparse Feature Coactivation Reveals Causal Semantic Modules in Large Language Models

Large language models (LLM) often hallucinate, and while adding citations is a common solution, it is frequently insufficient for accountability as users struggle to verify how a cited source supports a generated claim. Existing methods are typically coarse-grained and fail to distinguish between direct quotes and complex reasoning. In this paper, we introduce Generation-time Fine-grained Provenance, a task where models must generate fluent answers while simultaneously producing structured, sentence-level provenance triples. To enable this, we present ReFInE (Relation-aware Fine-grained Interpretability & Evidence), a dataset featuring expert-verified annotations that distinguish between Quotation, Compression, and Inference. Building on ReFInE, we propose GenProve, a framework that combines Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO). By optimizing a composite reward for answer fidelity and provenance correctness, GenProve significantly outperforms 14 strong LLMs in joint evaluation. Crucially, our analysis uncovers a reasoning gap where models excel at surface-level quotation but struggle significantly with inference-based provenance, suggesting that verifiable reasoning remains a frontier challenge distinct from surface-level citation.

GenProve: Learning to Generate Text with Fine-Grained Provenance

Reinforcement learning with verifiable rewards (RLVR) has significantly advanced the reasoning ability of vision-language models (VLMs). However, the inherent text-dominated nature of VLMs often leads to insufficient visual faithfulness, characterized by sparse attention activation to visual tokens. More importantly, our empirical analysis reveals that temporal visual forgetting along reasoning steps exacerbates this deficiency. To bridge this gap, we propose Visually-Guided Policy Optimization (VGPO), a novel framework to reinforce visual focus during policy optimization. Specifically, VGPO initially introduces a Visual Attention Compensation mechanism that leverages visual similarity to localize and amplify visual cues, while progressively elevating visual expectations in later steps to counteract visual forgetting. Building on this mechanism, we implement a dual-grained advantage re-weighting strategy: the intra-trajectory level highlights tokens exhibiting relatively high visual activation, while the inter-trajectory level prioritizes trajectories demonstrating superior visual accumulation. Extensive experiments demonstrate that VGPO achieves better visual activation and superior performance in mathematical multimodal reasoning and visual-dependent tasks. The code has been released at https://github.com/wzb-bupt/VGPO.

Visually-Guided Policy Optimization for Multimodal Reasoning

Audio chaptering, the task of automatically segmenting long-form audio into coherent sections, is increasingly important for navigating podcasts, lectures, and videos. Despite its relevance, research remains limited and text-based, leaving key questions unresolved about leveraging audio information, handling ASR errors, and transcript-free evaluation. We address these gaps through three contributions: (1) a systematic comparison between text-based models with acoustic features, a novel audio-only architecture (AudioSeg) operating on learned audio representations, and multimodal LLMs; (2) empirical analysis of factors affecting performance, including transcript quality, acoustic features, duration, and speaker composition; and (3) formalized evaluation protocols contrasting transcript-dependent text-space protocols with transcript-invariant time-space protocols. Our experiments on YTSeg reveal that AudioSeg substantially outperforms text-based approaches, pauses provide the largest acoustic gains, and current MLLMs struggle due to context limitations and weak instruction following.

Beyond Transcripts: A Renewed Perspective on Audio Chaptering

The deployment of Large Language Models (LLMs) on resource-constrained edge devices is increasingly hindered by prohibitive memory and computational requirements. While ternary quantization offers a compelling solution by reducing weights to {-1, 0, +1}, current implementations suffer from a fundamental misalignment with commodity hardware. Most existing methods must choose between 2-bit aligned packing, which incurs significant bit wastage, or 1.67-bit irregular packing, which degrades inference speed. To resolve this tension, we propose Sherry, a hardware-efficient ternary quantization framework. Sherry introduces a 3:4 fine-grained sparsity that achieves a regularized 1.25-bit width by packing blocks of four weights into five bits, restoring power-of-two alignment. Furthermore, we identify weight trapping issue in sparse ternary training, which leads to representational collapse. To address this, Sherry introduces Arenas, an annealing residual synapse mechanism that maintains representational diversity during training. Empirical evaluations on LLaMA-3.2 across five benchmarks demonstrate that Sherry matches state-of-the-art ternary performance while significantly reducing model size. Notably, on an Intel i7-14700HX CPU, our 1B model achieves zero accuracy loss compared to SOTA baselines while providing 25% bit savings and 10% speed up. The code is available at https://github.com/Tencent/AngelSlim.

Sherry: Hardware-Efficient 1.25-Bit Ternary Quantization via Fine-grained Sparsification

For Large Language Models (LLMs) to be reliably deployed, models must effectively know when not to answer: abstain. Reasoning models, in particular, have gained attention for impressive performance on complex tasks. However, reasoning models have been shown to have worse abstention abilities. Taking the vulnerabilities of reasoning models into account, we propose our Query Misalignment Framework. Hallucinations resulting in failed abstention can be reinterpreted as LLMs answering the wrong question (rather than answering a question incorrectly). Based on this framework, we develop a new class of state-of-the-art abstention methods called **Trace Inversion**. First, we generate the reasoning trace of a model. Based on only the trace, we then reconstruct the most likely query that the model responded to. Finally, we compare the initial query with the reconstructed query. Low similarity score between the initial query and reconstructed query suggests that the model likely answered the question incorrectly and is flagged to abstain. Extensive experiments demonstrate that **Trace Inversion** effectively boosts abstention performance in four frontier LLMs across nine abstention QA datasets, beating competitive baselines in 33 out of 36 settings. The code is available at this https://anonymous.4open.science/r/trace-inversion-08BB/.

Answering the Wrong Question: Reasoning Trace Inversion for Abstention in LLMs

As LLM-generated text is increasingly used, especially in fictional domains, we explore how much LLM-generated stories differ from human-written stories. In this work, we focus on characters. We borrow definitions from narratology to analyze 8 intricate category-pairs of character, such as stylization and wholeness. These category-pairs consider more than just basic characteristics. They assess how characters are portrayed within their stories. After automatically inferring categories of characters within both LLM and human-written stories, we compare and contrast these two sets of stories. We consider the following overarching questions: (1) Do LLMs and human-written stories have similar characters? and (2) Do LLMs generate stories with a variety of characters? Our analysis includes research questions that focus on stories generated by popular LLMs and recently published human-written stories. We describe a number of interesting similarities, differences and key takeaways.

CASPER in the Machine: Insights into Character Variety in LLM-Generated Stories

Recent evaluations show that large language models (LLMs) frequently fail to challenge users' harmful beliefs in domains ranging from medical advice to social reasoning. We present a unifying analysis through the lens of pragmatics: these safety failures can be understood and addressed as LLMs exhibiting excessive accommodation and insufficient epistemic vigilance. We show that the pragmatic factors affecting accommodation and epistemic vigilance in humans (at-issueness, linguistic encoding, and source reliability) influence LLM behaviors in similar ways. We demonstrate how these factors explain performance differences across three safety benchmarks that test models' ability to challenge harmful beliefs, spanning misinformation (Cancer-Myth, SAGE-Eval) and sycophancy (ELEPHANT). This pragmatic lens further motivates prompting interventions, such as adding the phrase "wait a minute", that drastically improve performance on these difficult benchmarks by shifting pragmatic cues. Our results have practical implications for benchmark design and underscore the importance of pragmatics for understanding model behavior and improving performance.

Accommodation and Epistemic Vigilance: A Pragmatic Account of Why LLMs Fail to Challenge Harmful Beliefs

AI guardrail systems support usage policies by determining whether a user query or a generated response is allowed or forbidden under the policy. Fine-tuned guardrails -- such as LlamaGuard and ShieldGemma -- include policy definitions in prompts during training that can be updated during inference to aid generalization. However, our analysis reveals that these models still overfit the training policies, which prevents adaptation to new domains. We propose Augmented Policy Training (APT), a training recipe that enhances guardrail adaptability to unseen policies by using a suite of policy perturbation strategies during training to reduce overfitting and increase generalization. Notably, a small 1B model trained in this manner achieves comparable or better performance than existing 8B guardrails on unseen policies. Our work reveals critical limitations of existing AI guardrails, offers a promising solution, and provides actionable insights for adapting systems to new domains and policies.

Domain Generalizable AI Guardrails with Augmented Policy Training

Conversational AI is increasingly used in e-commerce to deliver personalized customer support. We present a production RAG-based How-To Assistant that answers support and how-to queries by grounding responses in a proprietary knowledge base. We study three factors that drive quality: (1) document chunking and contextualization for indexing, (2) query refinement methods, and (3) automatic LLM-based evaluation for rapid iteration and reliable measurement. We also describe the end-to-end system workflow - from offline indexing to real-time serving and report deployment metrics, offering practical guidance for building scalable, high-precision RAG assistants in commercial support settings.

Downloads

Next from ACL 2026

Sparse Feature Coactivation Reveals Causal Semantic Modules in Large Language Models

Similar lecture

Don’t Forget About Pronouns: Removing Gender Bias in Language Models Without Losing Factual Gender Information

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ACL 2026

Sparse Feature Coactivation Reveals Causal Semantic Modules in Large Language Models

Similar lecture

Don’t Forget About Pronouns: Removing Gender Bias in Language Models Without Losing Factual Gender Information

Downloads