Austria

Recent advancements in generative large language models (LLMs) have enabled wider applicability, accessibility, and flexibility. However, their reliability and trustworthiness are still in doubt, especially for concerns regarding individuals&#39; data privacy. Great efforts have been made on privacy by building various evaluation benchmarks to study LLMs&#39; privacy awareness and robustness from their generated outputs to their hidden representations. Unfortunately, most of these works adopt a narrow formulation of privacy and only investigate personally identifiable information (PII). In this paper, we follow the merit of the Contextual Integrity (CI) theory, which posits that privacy evaluation should not only cover the transmitted attributes but also encompass the whole relevant social context through private information flows. We present PrivaCI-Bench, a comprehensive contextual privacy evaluation benchmark targeted at legal compliance to cover well-annotated privacy and safety regulations, real court cases, privacy policies, and synthetic data built from the official toolkit to study LLMs&#39; privacy and safety compliance. We evaluate the latest LLMs, including the recent reasoner models QwQ-32B and Deepseek R1. Our experimental results suggest that though LLMs can effectively capture key CI parameters inside a given context, they still require further advancements for privacy compliance.

ACL 2025

PrivaCI-Bench: Evaluating Privacy with Contextual Integrity and Legal Compliance

contextual integrity

safety

large language models

privacy

### Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Welcome to the 2025 meeting of the Nations of the Americas Chapter of the Association for Computational Linguistics! I am proud to help organize the first NAACL conference to carry the new name of our organization, one that emphasizes inclusion for all of the Americas. I am also pleased to welcome you to Albuquerque, New Mexico, a state whose unique blend of cultural influences will make for an excellent backdrop for NAACL 2025, especially with this year’s special theme on NLP in a Multicultural World. 
**[Continue reading...](https://drive.google.com/file/d/1jX-qGhqVSZZCIrAnJaz798pflu5Irrdn/view?usp=sharing)**

*- Colin Cherry, Google, NAACL 2025 General Chair* 

[![](https://assets.underline.io/markdown_image/1/image/b087f8a4dc5816d6e1a6514e59c59ac3.png)](https://drive.google.com/file/d/1T96GzPqObXrMln2BMByCSXTSizjTg69P/view?usp=sharing)

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

To access NAACL 2025 event page you are required to register. Please follow [**this link**](https://2025.naacl.org/registration/registration/) to register. Access will depend on your registration type.

Please register!

NAACL 2025

Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Recent advancements in generative large language models (LLMs) have enabled wider applicability, accessibility, and flexibility. However, their reliability and trustworthiness are still in doubt, especially for concerns regarding individuals' data privacy. Great efforts have been made on privacy by building various evaluation benchmarks to study LLMs' privacy awareness and robustness from their generated outputs to their hidden representations. Unfortunately, most of these works adopt a narrow formulation of privacy and only investigate personally identifiable information (PII). In this paper, we follow the merit of the Contextual Integrity (CI) theory, which posits that privacy evaluation should not only cover the transmitted attributes but also encompass the whole relevant social context through private information flows. We present PrivaCI-Bench, a comprehensive contextual privacy evaluation benchmark targeted at legal compliance to cover well-annotated privacy and safety regulations, real court cases, privacy policies, and synthetic data built from the official toolkit to study LLMs' privacy and safety compliance. We evaluate the latest LLMs, including the recent reasoner models QwQ-32B and Deepseek R1. Our experimental results suggest that though LLMs can effectively capture key CI parameters inside a given context, they still require further advancements for privacy compliance.

poster

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s 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.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

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 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Rich user-item interactions are essential for building reliable recommender systems, as they reflect user preference patterns. However, minor language news recommendation platforms suffer from limited interactions due to a small user base. A natural solution is to apply well-established English recommender systems to minor language news recommendation, but the linguistic gap can lead to inaccurate modeling of minor language news content. Therefore, enabling few-shot minor language news recommender systems to capture both content information and preference patterns remains a challenge. Based on the observation that preference patterns are similar across languages, we propose a minor language news recommendation model by cross-lingual preference pattern transfer, named PPT. Our model adopts the widely used two-tower architecture and employs the large language model as the backbone of the news encoder. Through cross-lingual alignment, the strong English capability of the news encoder is extended to minor languages, thus enhancing news content representations. Additionally, through cross-lingual news augmentation, PPT simulates interactions of minor language news in the English domain, which facilitates the transfer of preference patterns from the many-shot English domain to the few-shot minor language domain. Extensive experiments on two real-world datasets across 15 minor languages demonstrate the superiority and generalization of our proposed PPT in addressing minor language news recommendation.

PPT: A Minor Language News Recommendation Model via Cross-Lingual Preference Pattern Transfer

The widespread usage of online Large Language Models (LLMs) inference services has raised significant privacy concerns about the potential exposure of private information in user inputs. Existing privacy protection methods for LLMs suffer from either insufficient privacy protection with performance degradation, or large inference time overhead. To address these limitations, we propose PrivacyRestore, a plug-and-play method to protect the privacy of user inputs during LLM inference for the client-server scenario. The server first trains restoration vectors for each privacy span type offline and then releases them to the clients. During inference, the client aggregates restoration vectors of all privacy spans in the user query into a meta restoration vector, which is later sent to the server to restore information. Before transmission, the client removes all privacy spans in the user query and applies $d_\chi$-privacy mechanism to the meta vector for privacy protection. We prove that our method can inherently prevent the linear growth of the privacy budget. We conduct extensive experimental, covering the medical and legal domains, and demonstrate that PrivacyRestore effectively protects private information and maintains acceptable levels of performance and inference efficiency

PrivacyRestore: Privacy-Preserving Inference in Large Language Models via Privacy Removal and Restoration

Recent Continual Learning (CL)-based Temporal Knowledge Graph Reasoning (TKGR) methods focus on significantly reducing computational cost and mitigating catastrophic forgetting caused by fine-tuning models with new data. However, existing CL-based TKGR methods still face two key limitations: (1) They usually one-sidedly reorganize individual historical facts, while overlooking the historical context essential for accurately understanding the historical semantics of these facts; (2) They preserve historical knowledge by simply replaying historical facts, while ignoring the potential conflicts between historical and emerging facts. In this paper, we propose a $\textbf{D}$eep $\textbf{G}$enerative $\textbf{A}$daptive $\textbf{R}$eplay (DGAR) method, which can generate and adaptively replay historical entity distribution representations from the whole historical context. To address the first challenge, historical context prompts as sampling units are built to preserve the whole historical context information. To overcome the second challenge, a pre-trained diffusion model is adopted to generate the historical distribution. During the generation process, the common features between the historical and current distributions are enhanced under the guidance of the TKGR model. In addition, a layer-by-layer adaptive replay mechanism is designed to effectively integrate historical and current distributions. Experimental results demonstrate that DGAR significantly outperforms baselines in reasoning and mitigating forgetting.

A Generative Adaptive Replay Continual Learning Model for Temporal Knowledge Graph Reasoning

Predicting the types and affinities of protein-protein interactions (PPIs) is crucial for understanding biological processes and developing novel therapeutic approaches. While encoding proteins themselves is essential, PPI networks can also provide rich prior knowledge for these predictive tasks. However, existing methods oversimplify the problem of PPI prediction in a semi-supervised manner when utilizing PPI networks, limiting their practical application. Furthermore, how to effectively use the rich prior knowledge of PPI networks for novel proteins not present in the network remains an unexplored issue. Additionally, due to inflexible architectures, most of existing methods cannot handle complexes containing an flexible number of proteins. To overcome these limitations, we introduce LLaPA (Large Language and Protein Assistant), a multimodal large language model that integrates proteins and PPI networks. LLaPA offers a more rational approach to utilizing PPI networks for PPI prediction and can fully exploit the information of PPI networks for unseen proteins. Through natural language instructions, LLaPA can accept flexible number of protein sequences and has the potential to perform various protein tasks. Experiments show that LLaPA achieves state-of-the-art performance in multi-label PPI (mPPI) type prediction and is capable of predicting the binding affinity between multiple interacting proteins based on sequence data.

Large Language and Protein Assistant for Protein-Protein Interactions Prediction

In the interaction between agents and their environments, agents expand their capabilities by planning and executing actions. However, LLM-based agents face substantial challenges when deployed in novel environments or required to navigate unconventional action spaces. To empower agents to autonomously explore environments, optimize workflows, and enhance their understanding of actions, we propose SynWorld, a framework that allows agents to synthesize possible scenarios with multi-step action invocation within the action space and perform Monte Carlo Tree Search (MCTS) exploration to effectively refine their action knowledge in the current environment. Our experiments demonstrate that SynWorld is an effective and general approach to learning action knowledge in new environments.

SynWorld: Virtual Scenario Synthesis for Agentic Action Knowledge Refinement

Large language models (LLMs) have been widely deployed as autonomous agents capable of following user instructions and making decisions in real-world applications. Previous studies have made notable progress in benchmarking the instruction following capabilities of LLMs in general domains, with a primary focus on their inherent commonsense knowledge. Recently, LLMs have been increasingly deployed as domain-oriented agents, which rely on domain-oriented guidelines that may conflict with their commonsense knowledge. These guidelines exhibit two key characteristics: they consist of a wide range of domain-oriented rules and are subject to frequent updates. Despite these challenges, the absence of comprehensive benchmarks for evaluating the domain-oriented guideline following capabilities of LLMs presents a significant obstacle to their effective assessment and further development. In this paper, we introduce GuideBench, a comprehensive benchmark designed to evaluate guideline following performance of LLMs. GuideBench evaluates LLMs on three critical aspects: (i) adherence to diverse rules, (ii) robustness to rule updates, and (iii) alignment with human preferences. Experimental results on a range of LLMs indicate substantial opportunities for improving their ability to follow domain-oriented guidelines. Data and code are available at Anonymous.

GuideBench: Benchmarking Domain-Oriented Guideline Following for LLM Agents

Mainstream issue-resolving frameworks predominantly rely on commercial models, leading to high costs and privacy concerns. Existing training approaches for issue resolving struggle with poor generalization and fail to fully leverage open-source development resources. We propose **S**ubtask-**o**riented **R**einforced **F**ine-**T**uning (**SoRFT**), a novel training approach to enhance the issue resolving capability of LLMs. We decomposes issue resolving into structured subtasks: file localization, function localization, line localization, and code edit generation. SoRFT consists of two training stages: (1) **rejection-sampled supervised fine-tuning**, Chain of Thought (CoT) data is filtered using ground-truth before fine-tuning the LLM, and (2) **rule-based reinforcement learning**, which leverages PPO with ground-truth based rewards. We evaluate the SoRFT-trained model on SWE-Bench Verified and SWE-Bench Lite, achieving state-of-the-art (SOTA) performance among open-source models (e.g., resolve 21.4% issues on SWE-Bench Verified with SoRFT-Qwen-7B). The experimental results demonstrate that SoRFT significantly enhances issue-resolving performance, improves model generalization, and provides a cost-efficient alternative to commercial models.

SoRFT: Issue Resolving with Subtask-oriented Reinforced Fine-Tuning

Automatic exploit generation (AEG) refers to the automatic discovery and exploitation of vulnerabilities against unknown targets. Traditional AEG often targets a single type of vulnerability and still relies on templates built from expert experience. To achieve intelligent exploit generation, we establish a comprehensive benchmark using Binary Exploitation (pwn) challenges in Capture the Flag (CTF) competitions and investigate the capabilities of Large Language Models (LLMs) in AEG based on the benchmark. To improve the performance of AEG, we propose PwnGPT, an LLM-based automatic exploit generation framework that automatically solves pwn challenges. The structural design of PwnGPT is divided into three main components: analysis, generation, and verification modules. With the help of a modular approach and structured problem inputs, PwnGPT can solve challenges that LLMs cannot directly solve. We evaluate PwnGPT on our benchmark and analyze the outputs of each module. Experimental results show that our framework is highly autonomous and capable of addressing various challenges. Compared to direct input LLMs, PwnGPT increases the completion rate of exploit on our benchmark from 26.3\% to 57.9\% with the OpenAI o1-preview model and from 21.1\% to 36.8\% with the GPT-4o model.

PwnGPT: Automatic Exploit Generation Based on Large Language Models

While data synthesis and distillation are promising strategies to enhance small language models, current approaches heavily rely on Large Language Models (LLMs), which suffer from high computational costs, environmental inefficiency, and potential biases inherited from monolithic architectures. In contrast, smaller LMs are more accessible and sustainable, but their individual capabilities often fall short in generating high-quality, diverse, and reliable data. Inspired by collaborative human processes (e.g., peer review), we propose a multiple small LMs involved framework, GRA, that aggregates specialized roles across small LMs to iterative refinement and quality control typically achieved by a single large LM. 
In this collaborative framework, multiple small LMs assume distinct roles—Generator, Reviewer, and Adjudicator—to simulate a peer-review-inspired data synthesis pipeline. The Generator proposes initial data samples, the Reviewer critiques their quality and diversity, and the Adjudicator resolves conflicts to finalize the output. By decomposing the synthesis process into specialized sub-tasks, collaborative small LMs can achieve data-level parity with distillation from large LMs. Through experiments across multiple benchmarks, we demonstrate that GRA-produced data matches or exceeds the quality of single large LM outputs, e.g., Qwen-2.5-72B-Instruct. Our results challenge the necessity of monolithic large models for high-quality data synthesis, advocating instead for strategic coordination of smaller agents.

A Strategic Coordination Framework of Small LMs Matches Large LMs in Data Synthesis

We introduce PaSa, an advanced Paper Search agent powered by large language models. PaSa can autonomously make a series of decisions, including invoking search tools, reading papers, and selecting relevant references, to ultimately obtain comprehensive and accurate results for complex scholar queries. We optimize PaSa using reinforcement learning with a synthetic dataset, AutoScholarQuery, which includes 35k fine-grained academic queries and corresponding papers sourced from top-tier AI conference publications. Additionally, we develop RealScholarQuery, a benchmark collecting real-world academic queries to assess PaSa performance in more realistic scenarios. Despite being trained on synthetic data, PaSa significantly outperforms existing baselines on RealScholarQuery, including Google, Google Scholar, Google with GPT-4o for paraphrased queries, ChatGPT (search-enabled GPT-4o), GPT-o1, and PaSa-GPT-4o (PaSa implemented by prompting GPT-4o). Notably, PaSa-7B surpasses the best Google-based baseline, Google with GPT-4o, by 37.78% in recall@20 and 39.90% in recall@50, and exceeds PaSa-GPT-4o by 30.36% in recall and 4.25% in precision. Model, datasets, and code are available at https://github.com/bytedance/pasa.
Demo: https://pasa-agent.ai

Downloads

Next from ACL 2025

PPT: A Minor Language News Recommendation Model via Cross-Lingual Preference Pattern Transfer

Similar lecture

Privacy Checklist: Privacy Violation Detection Grounding on Contextual Integrity Theory