Austria

Large Language Models (LLMs) are increasingly used in high-stakes domains like law and research, yet their inconsistencies and response instability raise concerns about trustworthiness. This study evaluates six leading LLMs—GPT-3.5, GPT-4, Claude, Gemini, Mistral, and LLaMA 2—on rationality, stability, and ethical fairness through reasoning tests, legal challenges, and bias-sensitive scenarios. Results reveal significant inconsistencies, highlighting trade-offs between model scale, architecture, and logical coherence. These findings underscore the risks of deploying LLMs in legal and policy settings, emphasizing the need for AI systems that prioritize transparency, fairness, and ethical robustness.

ACL 2025

The Lawyer That Never Thinks: Consistency and Fairness as Keys to Reliable AI

multistate bar examination (mbe)

intraclass correlation coefficient (icc)

ai evaluation and auditing

response variability

test-retest consistency

### 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)

technical paper

Lexical relation refers to the way words are related within a language. Prior work has demonstrated that pretrained language models (PLMs) can effectively mine lexical relations between word pairs. However, they overlook the potential of graph structures composed of lexical relations, which can be integrated with the semantic knowledge of PLMs. In this work, we propose a parameter-efficient fine-tuning method through graph context, which integrates graph features and semantic representations for lexical relation classification (LRC) and lexical entailment (LE) tasks. Our experiments show that graph features can help PLMs better understand more complex lexical relations, establishing a new state-of-the-art for LRC and LE. Finally, we perform an error analysis, identifying the bottlenecks of language models in lexical relation mining tasks and providing insights for future improvements.

Introducing Graph Context into Language Models through Parameter-Efficient Fine-Tuning for Lexical Relation Mining

With the increasing capability of large language models (LLMs), LLM-as-a-judge has emerged as a new evaluation paradigm. Compared with traditional automatic and manual evaluation, LLM evaluators exhibit better interpretability and efficiency. Despite this, existing LLM evaluators suffer from limited use scenarios and poor flexibility. To mitigate these issues, we propose Praetor, a fine-grained generative LLM evaluator with instance-level customazable evaluation criteria. To train Praetor, we curate a large-scale dataset guided with a hierarchical guideline covering a wide range of tasks and instance-level evaluation criteria. We train Praetor on this dataset in a multi-task learning fashion, which enables to evaluate LLMs in either pointwise grading or pairwise comparison way and support two languages simultaneously with a high flexibility of setting evaluation criteria. Extensive experiments demonstrate that Praetor outperforms previous LLM evaluators and instruction-tuned LLMs on multiple benchmarks, setting new SOTA results. It also exhibits the potential for generating critiques as scalable feedback to further improve LLMs. Our model and related resources are released at \url{https://github.com/tjunlp-lab/Praetor}.

Praetor: A Fine-Grained Generative LLM Evaluator with Instance-Level Customizable Evaluation Criteria

Tree of Thoughts (ToT) enhances Large Language Model (LLM) reasoning by structuring problem-solving as a spanning tree. However, recent methods focus on search accuracy while overlooking computational efficiency. The challenges of accelerating the ToT lie in the frequent switching of reasoning focus, and the redundant exploration of suboptimal solutions. To alleviate this dilemma, we propose Dynamic Parallel Tree Search (DPTS), a novel parallelism framework that aims to dynamically optimize the reasoning path in inference. 
It includes the Parallelism Streamline in the generation phase to build up a flexible and adaptive parallelism with arbitrary paths by cache management and alignment. Meanwhile, the Search and Transition Mechanism filters potential candidates to dynamically maintain the reasoning focus on more possible solutions with less redundancy. Experiments on Qwen-2.5 and Llama-3 on math and code datasets show that DPTS significantly improves efficiency by 2-4$\times$ on average while maintaining or even surpassing existing reasoning algorithms in accuracy, making ToT-based reasoning more scalable and computationally efficient. Codes are released at: https://github.com/yifu-ding/DPTS.

Dynamic Parallel Tree Search for Efficient LLM Reasoning

Long Context Understanding (LCU) is a critical area for exploration in current large language models (LLMs). However, due to the inherently lengthy nature of long-text data, existing LCU benchmarks for LLMs often result in prohibitively high evaluation costs, like testing time and inference expenses. Through extensive experimentation, we discover that existing LCU benchmarks exhibit significant redundancy, which means the inefficiency in evaluation. In this paper, we propose a concise data compression method tailored for long-text data with sparse information characteristics. By pruning the well-known LCU benchmark LongBench, we create MiniLongBench. This benchmark includes only 237 test samples across six major task categories and 21 distinct tasks. Through empirical analysis of over 60 LLMs, MiniLongBench achieves an average evaluation cost reduced to only 4.5% of the original while maintaining an average rank correlation coefficient of 0.97 with LongBench results. Therefore, our MiniLongBench, as a low-cost benchmark, holds great potential to substantially drive future research into the LCU capabilities of LLMs.

MiniLongBench: The Low-cost Long Context Understanding Benchmark for Large Language Models

Large language models (LLMs) augmented with retrieval systems have significantly advanced natural language processing tasks by integrating external knowledge sources, enabling more accurate and contextually rich responses. To improve the robustness of such systems against noisy retrievals, Retrieval-Augmented Fine-Tuning (RAFT) has emerged as a widely adopted method. However, RAFT conditions models to generate answers even in the absence of reliable knowledge. This behavior undermines their reliability in high-stakes domains, where acknowledging uncertainty is critical. To address this issue, we propose Divide-Then-Align (DTA), a post-training approach designed to endow RAG systems with the ability to respond with "I don't know" when the query is out of the knowledge boundary of both the retrieved passages and the model's internal knowledge. DTA divides data samples into four knowledge quadrants and constructs tailored preference data for each quadrant, resulting in a curated dataset for Direct Preference Optimization (DPO). Experimental results on three benchmark datasets demonstrate that \ourmethod effectively balances accuracy with appropriate abstention, enhancing the reliability and trustworthiness of retrieval-augmented systems.

Divide-Then-Align: Honest Alignment based on the Knowledge Boundary of RAG

Large Vision-Language Models (LVLMs) have demonstrated impressive multimodal abilities but remain prone to multilingual object hallucination, with a higher likelihood of generating responses inconsistent with the visual input when utilizing queries in non-English languages compared to English. Most existing approaches to address these rely on pretraining or fine-tuning, which are resource-intensive. In this paper, inspired by observing the disparities in cross-modal attention patterns across languages, we propose Cross-Lingual Attention Intervention for Mitigating multilingual object hallucination (CLAIM) in LVLMs, a novel near training-free method by aligning attention patterns. CLAIM first identifies language-specific cross-modal attention heads, then estimates language shift vectors from English to the target language, and finally intervenes in the attention outputs during inference to facilitate cross-lingual visual perception capability alignment. Extensive experiments demonstrate that CLAIM achieves an average improvement of 13.56% (up to 30% in Spanish) on the POPE and 21.75% on the hallucination subsets of the MME benchmark across various languages. Further analysis reveals that multilingual attention divergence is most prominent in intermediate layers, highlighting their critical role in multilingual scenarios.

CLAIM: Mitigating Multilingual Object Hallucination in Large Vision-Language Models with Cross-Lingual Attention Intervention

Large Language Models (LLMs) are increasingly used to assess NLP tasks due to their ability to generate human-like judgments. Single LLMs were used initially, however, recent work suggests using multiple LLMs as judges yields improved performance. An important step in exploiting multiple judgements is the combination stage, aggregation. Existing methods in NLP either assign equal weight to all LLM judgments or are designed for specific tasks such as hallucination detection. This work focuses on aggregating predictions from multiple systems where no reference labels are available. A new method called SkillAggregation is proposed, which learns to combine estimates from LLM judges without needing additional data or ground truth. It extends the Crowdlayer aggregation method, developed for image classification, to exploit the judge estimates during inference. The approach is compared to a range of standard aggregation methods on HaluEval-Dialogue, TruthfulQA and Chatbot Arena tasks. SkillAggregation outperforms Crowdlayer on all tasks, and yields the best performance over all approaches on the majority of tasks.

SkillAggregation: Reference-free LLM-Dependent Aggregation

In this work, we investigate how to sparsify a pre-trained dense large language model into a mixture-of-experts (MoE) architecture for faster inference. Our approach applies mask matrix to the activations for each expert, constrained by $L_0$ regularization to minimize the number of activated parameters. Starting with all parameters active, the model is progressively sparsified during training, ensuring minimal performance loss. This approach proves more efficient than one-shot sparsification techniques, which typically require significant resources for performance recovery. Moreover, our approach automatically identifies shared, token-specific, and inactive experts, allowing for more efficient allocation of computational resources. Through extensive experiments, we achieve up to 97% performance retention on downstream tasks with only 50% of the feed-forward parameters activated in dense models. Beyond enhancing inference efficiency, this strategy of sharing computational units among experts presents a valuable framework for designing more generalized and efficient MoE architectures, opening avenues for future advancements in expert-based models.

Masks Can be Learned As An Alternative of Experts

It is well-known that a diverse corpus is critical for training large language models, which are typically constructed from a mixture of various domains. In general, previous efforts resort to either sampling training data from different domains with static proportions or dynamically adjusting these proportions during training to optimise pretraining performance. However, few methods addressed the complexities of domain-adaptive continual pre-training. To fill this gap, we propose Velocitune, a novel framework that dynamically assesses learning velocity and adjusts data proportions accordingly, favouring slower learning domains while de-emphasising faster learning ones, which is guided by a scaling law to estimate the desired learning goal for each domain with a less associated cost. To evaluate the effectiveness of Velocitune, we conduct experiments on a dataset focused on reasoning tasks with CodeLlama, as well as on a corpus of system commands using Llama3 and Mistral. Velocitune achieves performance gains in both math and code reasoning tasks and command-line generation benchmarks. Further analysis reveals that key factors driving Velocitune's effectiveness include target estimation and data ordering.

Velocitune: A Velocity-based Dynamic Domain Reweighting Method for Continual Pre-training

Fine-Tuning of LLMs using RLHF / RLAIF has been shown as a critical step to improve the performance of LLMs in complex generation tasks. These methods typically use response-level human or model feedback for alignment. Recent works indicate that finer sentence or span-level labels provide more accurate and interpretable feedback for LLM optimization. In this work, we propose FRACTAL, a suite of models to disaggregate response-level labels into sentence-level (pseudo-)labels through Multiple Instance Learning (MIL) and Learning from Label Proportions (LLP) formulations, novel usage of prior information, and maximum likelihood calibration. We perform close to 2000 experiments across 6 datasets and 4 tasks that show that FRACTAL can reach up to 93% of the performance of the fully supervised baseline while requiring only around 10% of the gold labels. Furthermore, in a downstream eval, employing step-level pseudo scores in RLHF for a math reasoning task leads to 5% absolute improvement in performance. Our work is the first to develop response-level feedback to sentence-level scoring techniques leveraging sentence-level prior information, along with comprehensive evaluations on multiple tasks as well as end-to-end finetuning evaluations.

Downloads

Next from ACL 2025

Introducing Graph Context into Language Models through Parameter-Efficient Fine-Tuning for Lexical Relation Mining

Similar lecture

The Impossibility of Fair LLMs