United States

Large language models (LLMs) for code rely on subword tokenizers, such as byte-pair encoding (BPE), learned from mixed natural language text and programming language code but driven by statistics rather than grammar. As a result, semantically identical code snippets can be tokenized differently depending on superficial factors such as whitespace or identifier naming. To measure the impact of this misalignment, we introduce TokDrift, a framework that applies semantic-preserving rewrite rules to create code variants differing only in tokenization. Across nine code LLMs, including large ones with over 30B parameters, even minor formatting changes can cause substantial shifts in model behavior. Layer-wise analysis shows that the issue originates in early embeddings, where subword segmentation fails to capture grammar token boundaries. Our findings identify misaligned tokenization as a hidden obstacle to reliable code understanding and generation, highlighting the need for grammar-aware tokenization for future code LLMs.

ACL 2026

TokDrift: When LLM Speaks in Subwords but Code Speaks in Grammar

### 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**. 

ACL 2025

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

poster

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.

**Online Registration Form**: https://acl.swoogo.com/acl2026

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.

Retrieval-augmented generation (RAG) enables large language models (LLMs) to dynamically access external information, which is powerful for answering questions over previously unseen documents. Nonetheless, they struggle with high-level conceptual understanding and holistic comprehension due to limited context windows, which constrain their ability to perform deep reasoning over long-form, domain-specific content such as full-length books. To solve this problem, knowledge graphs (KGs) have been leveraged to provide entity-centric structure and hierarchical summaries, offering more structured support for reasoning. However, existing KG-based RAG solutions remain restricted to text-only inputs and fail to leverage the complementary insights provided by other modalities such as vision. On the other hand, reasoning from visual documents requires textual, visual, and spatial cues into structured, hierarchical concepts. To address this issue, we introduce a multimodal knowledge graph-based RAG that enables cross-modal reasoning for better content understanding. Our method incorporates visual cues into the construction of knowledge graphs, the retrieval phase, and the answer generation process. Experimental results across both global and fine-grained question answering tasks show that our approach consistently outperforms existing approaches on both textual and multimodal benchmarks.

MegaRAG: Multimodal Knowledge Graph-Based Retrieval Augmented Generation

Large language models have seen widespread adoption, yet they remain vulnerable to multi-turn jailbreak attacks, threatening their safe deployment. This has led to the task of training automated multi-turn attackers to probe model safety vulnerabilities. However, existing approaches typically rely on turn-level optimization, which is insufficient for learning long-term attack strategies. To bridge this gap, we formulate this task as a multi-turn reinforcement learning problem, directly optimizing the harmfulness of the final-turn response as the outcome reward. To address the sparse supervision of the outcome reward, we introduce TROJail, which employs two process rewards to evaluate the utility of intermediate prompts and integrate them into advantage estimation. These rewards (1) penalize overly harmful prompts that trigger the model’s refusal mechanism, and (2) encourage steering the semantic relevance of responses toward the targeted harmful content. Experimental results show improved attack success rates across multiple models and benchmarks, highlighting the effectiveness of our approach. The code is available at \url{https://anonymous.4open.science/r/TROJail}. \textcolor{red}{Warning: This paper contains examples of harmful content.}

TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards

We introduce MinerU2.5, a 1.2B-parameter document parsing vision-language model that achieves state-of-the-art recognition accuracy while maintaining exceptional computational efficiency. Our approach employs a coarse-to-fine, two-stage parsing strategy that decouples global layout analysis from local content recognition. In the first stage, the model performs efficient layout analysis on downsampled images to identify structural elements, circumventing the computational overhead of processing high-resolution inputs. In the second stage, guided by the global layout, it performs targeted content recognition on native-resolution crops extracted from the original image, preserving fine-grained details in dense text, complex formulas, and tables. To support this strategy, we developed a comprehensive data engine that generates diverse, large-scale training corpora for both pretraining and fine-tuning. Ultimately, MinerU2.5 demonstrates strong document parsing ability, achieving state-of-the-art performance on multiple benchmarks, surpassing both general-purpose and domain-specific models across various recognition tasks, while maintaining significantly lower computational overhead.

MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing

Misinformation on social media undermines informed decision-making and public trust. Prebunking offers a proactive complement by helping users recognize manipulation tactics before they encounter them in the wild. We present **CritiSense**, a mobile media-literacy app that builds these skills through short, interactive challenges with instant feedback. It is the **first** multilingual (supporting nine languages) and modular platform, designed for rapid updates across topics and domains. We report a usability study with 93 users: 83.9% expressed overall satisfaction and 90.1% rated the app as easy to use. Qualitative feedback indicates that **CritiSense** helps improve digital literacy skills. Overall, it provides a multilingual prebunking platform and a testbed for measuring the impact of microlearning on misinformation resilience. Over 3+ months, we have reached 300+ active users. It is freely available to all users on the Apple [App Store](https://apps.apple.com/us/app/critisense/id6749675792) and [Google Play Store](https://play.google.com/store/apps/details?id=com.critisense&hl=en). **Demo Video:** [https://shorturl.at/CDcdc](https://shorturl.at/CDcdc)

G-D1 CritiSense: Critical Digital Literacy and Resilience Against Misinformation

Large language models (LLMs) enable scalable content generation for personalized learning, but reliability and pedagogical alignment remain open challenges. We present PathBuilder, a web-based system that integrates expert-validated assessment, retrieval-augmented generation (RAG), and an LLM-as-a-Judge validation loop within a closed instructional pipeline. The system uses a 17,758-item curriculum-aligned question bank, including 1,018 expert-approved LLM-generated items, to construct diagnostic and post-tests for fine-grained learner profiling. In a real-world deployment with 179 registered users (75 matched learners), PathBuilder achieved a mean absolute gain of 37.9 percentage points, Hake’s normalized gain of 0.760, and a large effect size (Cohen’s d = 0.98). A controlled study of the judge mechanism showed consistent high-quality instructional outputs with a 100\% threshold pass rate. These results demonstrate that structured curriculum alignment combined with retrieval grounding and automated validation can support reliable LLM-based personalization in deployed learning systems. A live demonstration of PathBuilder is available at https://demo.pathbuilderedu.com.

G-D2 PathBuilder: A Quality-Controlled LLM System for Personalized Learning Pathways

Material claims in papers, patents, etc., often involve physical feasibility (e.g., stability under conditions, property consistency), not just textual feasibility. Yet most claim verifiers operate over language, therefore producing ungrounded judgments. On the other hand, direct
first-principles verification (e.g., density functional theory, DFT) is inflexible and hard to invoke from underspecified free-form claims.
Therefore, we introduce **PhyVer**, a **phy**sics-grounded material claim **ver**ification system that bridges this gap by translating claims
into multi-fidelity physical evidence and interpretable verdicts. To support human-in-the-loop inspection, we present an interactive web interface that visualizes the instantiated structure, optimization trajectories, DFT summaries, and the final decision. On expert-labeled claims, **PhyVer** improves agreement with experts over text-only GPT-5.1, reducing MAE from 1.54 to 1.20 and Signed MAE from
0.95 to 0.82, and increasing Accuracy@±1 from 50% to 70%.

G-D3 PhyVer: Physics-Grounded Material Claim Verification with Multi-Fidelity Physical Evidence

We present EduCoder, an open-source web platform designed for annotating classroom conversation transcripts. Existing annotation tools do not support the team-based workflows or access to instructional context that education discourse research requires. EduCoder addresses these gaps by combining transcript text, synchronized video, and instructional materials within a single workspace. The platform supports scoping annotation to specific portions of a lesson, coordinating work across annotation teams, and optionally integrating LLM-generated annotations with structured human–LLM comparison. EduCoder is freely accessible at https://edu-coder.com and demonstrated in an accompanying video (https://www.youtube.com/watch?v=L4MzvmylgW0).

G-D4 EduCoder: An Open-Source Annotation System for Education Transcript Data

We present **CPTCoder**, a human-in-the-loop system that predicts standardized medical procedure codes from clinical text. Clinical procedure coding is an extreme multi-label classification problem over a long-tailed space of short numeric identifiers, where a single-digit difference denotes an entirely different procedure. CPTCoder adapts an instruction-tuned LLM with a code-aware vocabulary and constrained decoding that guarantees all outputs are valid codes. To support human review, we derive per-code posterior inclusion probabilities from n-best reweighting, producing interpretable confidence scores that rank predictions and flag uncertain cases. A post-decoding constraint repair step enforces mutual-exclusion rules between conflicting codes.


To enable reproducible research in this underexplored setting, we release **MIMIC-CPT**, a PhysioNet-accessible benchmark of 37,885 expert-cleaned report–code pairs with a deliberately hardened test split: 88\% of test examples contain label combinations unseen during training, and over a third include codes with five or fewer training occurrences. We additionally provide 413,085 weakly aligned pairs and evaluate on a separate live dataset from a hospital, which includes out-of-domain radiology reports with billing-expert-verified labels. CPTCoder achieves 0.61 and 0.51 micro-F1 on the hardened MIMIC split and Hospital-298 respectively, outperforming the strongest baseline by 12 and 5 absolute points while reducing digit-level near-miss errors.

G-D5 CPTCoder: A Reliable LLM System for Medical Procedure Code Prediction

Understanding how developers interact with AI coding assistants requires more than chat logs or git histories in isolation; it requires reconstructing the full context: which prompt led to which edit, what the developer tried and discarded, and how their strategy evolved over time. We present RECAP (Replay, Embed, Classify AI Prompting), an open-source platform that (1) passively records AI chat sessions and fine-grained code edits inside VS Code without disrupting the developer's workflow, (2) merges them into a unified timeline for interactive session replay, and (3) provides extensible analysis modules for behavioral classification and AI reliance measurement. Deployed in a university software engineering course, RECAP captured 2,034 prompts and 8,239 code edits from 41 students across a multi-week project. We demonstrate how the platform's linked data and replay capabilities enable analyses of developer-AI interaction patterns that no single data source could support.

RECAP: An End-to-End Platform for Capturing, Replaying, and Analyzing AI-Assisted Programming Interactions

Test-time compute (TTC) scaling has emerged as a powerful paradigm for improving large language model (LLM) reasoning by allocating additional compute during inference, e.g., via multi-sample generation and verifier-based reranking. Existing TTC scaling strategies and reasoning scorers remain fragmented, evaluated under inconsistent protocols, and are rarely analyzed through the lens of quality-cost trade-offs. We introduce ThinkBooster, a unified framework for seamless test-time compute scaling of LLM reasoning, which consists of (i) a modular Python library implementing state-of-the-art TTC scaling strategy and scorer families, (ii) a benchmark that jointly evaluates performance and compute efficiency, and (iii) a deployable OpenAI-compatible proxy service that enables drop-in integration of adaptive reasoning into real-world applications. We further provide a demo visual debugger for inspecting the reasoning trajectories and escalation decisions. Empirical results on mathematical and coding tasks reveal the performance-compute trade-offs of TTC scaling strategies and scoring methods, and demonstrate that ThinkBooster provides practical gains in real-world tasks. The code is available online under an MIT license.

Downloads

Next from ACL 2026

MegaRAG: Multimodal Knowledge Graph-Based Retrieval Augmented Generation

Similar lecture

Grammar-Based Code Representation: Is It a Worthy Pursuit for LLMs?