China

We propose a new, training-free method, Graph Reasoning via Retrieval Augmented Framework (GRRAF), that harnesses retrieval-augmented generation (RAG) alongside the code-generation capabilities of large language models (LLMs) to address a wide range of graph reasoning tasks. In GRRAF, the target graph is stored in a graph database, and the LLM is prompted to generate executable code queries that retrieve the necessary information. This approach circumvents the limitations of existing methods that require extensive finetuning or depend on predefined algorithms, and it incorporates an error feedback loop with a time-out mechanism to ensure both correctness and efficiency. Experimental evaluations on the GraphInstruct dataset reveal that GRRAF achieves 100% accuracy on most graph reasoning tasks, including cycle detection, bipartite graph checks, shortest path computation, and maximum flow, while maintaining consistent token costs regardless of graph sizes. Imperfect but still very high performance is observed on subgraph matching. Notably, GRRAF scales effectively to large graphs with up to 10,000 nodes.

EMNLP 2025

Zero-shot Graph Reasoning via Retrieval Augmented Framework with LLMs

llm application

graph reasoning

### 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

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

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

The influence of personas on Large Language Models (LLMs) has been widely studied, yet their direct impact on performance remains uncertain. This work explores a novel approach to guiding LLM behaviour through role vectors, an alternative to persona-based prompting. We construct 29 role vectors derived from model activations and evaluate their impact on benchmark performance across multiple domains. Our analysis investigates whether these vectors can effectively steer models toward domain-specific expertise. We measure two key interventions: (i) activation addition, which reinforces role-specific directions, and (ii) directional ablation, which removes them. Results on well-established benchmarks indicate that role vectors do, in fact, influence model behaviour, improving task performance in relevant domains while marginally affecting unrelated tasks. This, in turn, suggests that manipulating internal model representations has a greater impact on outcomes than persona-based prompting.

Can Role Vectors Affect LLM Behaviour?

We investigate the effect of duplicating multihead self-attention (MHSA) layers in large language models (LLMs) across a range of language tasks, with and without fine-tuning. The results demonstrate that duplicating the initial layers once or twice often yields a significant performance boost. Attention analysis uncovered the underlying mechanisms driving the improvement when performing layer duplication. This method enhances LLM capabilities with or without additional training or labeled data.

Layer Duplication in LLMs

The past year has seen an increase in the number of large (parameter) mixture-of-expert (MoE) models in the open domain. With the increase in size has come an increase in capabilities, with the largest models achieving new state-of-the-art on mathematics and reasoning benchmarks. Although the benefit of the MoE architecture can be understood through an engineering efficiency lens, prior work has also investigated whether MoE models exhibit functional differentiation along architectural lines, by examining patterns in the routing behavior of MoE modules. This work investigates whether expert routing in large MoE models is related to the _semantics_ of the inputs. We develop a pair of controlled experiments to identify whether routing occurs on the basis of _semantics_, i.e. the meaning of the inputs. The first experiment compares the activations on lexically identical but semantically distinct inputs, by using pairs of sentences where a common target word either has the same meaning or not. The second experiment compares activations where context is kept constant but the target word is substituted for semantically equivalent/distinct words. By comparing the rate of overlap in expert routing between sentence pairs where meaning is kept the same versus sentence pairs where meaning is changed, we identify whether expert routing is made on the basis of semantics. Our experiments show clear and statistically significant evidence of _semantic routing_ in large MoE models; inputs with the same meaning have greater overlap in routed experts than inputs with differing meaning.

Probing Semantic Routing in Large Mixture-of-Expert Models

Digital tool-based agents that invoke external Application Programming Interfaces (APIs) often rely on documentation to understand API functionality. However, such documentation is frequently missing, outdated, privatized, or inconsistent—hindering the development of reliable, general-purpose agents. In this work, we propose learning API functionality directly from demonstrations as a new paradigm applicable in scenarios without documentation. Using existing API benchmarks, we collect demonstrations from both expert API-based agents and from self-exploration. To understand what information demonstrations must convey for successful task completion, we extensively study how the number of demonstrations and the use of LLM-generated summaries and evaluations affect the task success rate of the API-based agent. Our experiments across datasets and models show that learning functionality from demonstrations remains a non-trivial challenge, even for state-of-the-art LLMs. We find that providing explicit function calls and natural language critiques significantly improves the agent’s task success rate due to more accurate parameter filling. We analyze failure modes, identify sources of error, and highlight key open challenges for future work in documentation-free, self-improving, API-based agents.

Learning API Functionality from In-Context Demonstrations for Tool-based Agents

Recent work has investigated the capabilities of large language models (LLMs) as zero-shot models for generating individual-level characteristics (e.g., to serve as risk models or augment survey datasets). However, when should a user have confidence that an LLM will provide high-quality predictions for their particular task? To address this question, we conduct a large-scale empirical study of LLMs' zero-shot predictive capabilities across a wide range of tabular prediction tasks. We find that LLMs' performance is highly variable, both on tasks within the same dataset and across different datasets. However, when the LLM performs well on the base prediction task, its predicted probabilities become a stronger signal for individual-level accuracy. Then, we construct metrics to predict LLMs' performance at the task level, aiming to distinguish between tasks where LLMs may perform well and where they are likely unsuitable. We find that some of these metrics, each of which are assessed without labeled data, yield strong signals of LLMs' predictive performance on new tasks.

Predicting Language Models’ Success at Zero-Shot Probabilistic Prediction

A dangerous assumption that can be made from prior work on the bias transfer hypothesis (BTH) is that biases do not transfer from pre-trained large language models (LLMs) to adapted models. We invalidate this assumption by studying the BTH in causal models under prompt adaptations, as prompting is an extremely popular and accessible adaptation strategy used in real-world applications. In contrast to prior work, we find that biases can transfer through prompting and that popular prompt-based mitigation methods do not consistently prevent biases from transferring. Specifically, the correlation between intrinsic biases and those after prompt adaptation remained moderate to strong across demographics and tasks: gender (rho >= 0.94) in co-reference resolution, and for age (rho >= 0.98), religion (rho >= 0.69), etc., in question answering. Further, we find that biases remain strongly correlated when varying few-shot composition parameters, such as sample size, stereotypical content, occupational distribution and representational balance (rho >= 0.90). We evaluate several prompt-based debiasing strategies and find that different approaches have distinct strengths, but none consistently reduce bias transfer across models, tasks or demographics. These results demonstrate that correcting bias, and potentially improving reasoning ability, in intrinsic models may be reliable ways to prevent propagation of biases to downstream tasks.

Bias after Prompting: Persistent Discrimination in Large Language Models

Large Language Models (LLMs) typically rely on a large number of parameters for token embedding, leading to substantial storage requirements and memory footprints. In particular, LLMs deployed on edge devices are memory-bound, and reducing the memory footprint by compressing the embedding layer not only frees up the memory bandwidth but also speeds up inference. To address this, we introduce CARVQ, a post-training novel Corrective Adaptor combined with group Residual Vector Quantization. CARVQ relies on the composition of both linear and non-linear maps and mimics the original model embedding to compress to approximately 1.6 bits without requiring specialized hardware to support lower-bit storage. We test our method on pre-trained LLMs such as LLaMA-3.2-1B, LLaMA-3.2-3B, LLaMA-3.2-3B-Instruct, LLaMA-3.1-8B, Qwen2.5-7B, Qwen2.5-Math-7B and Phi-4, evaluating on common generative, discriminative, math and reasoning tasks. We show that in most cases, CARVQ can achieve lower average bitwidth-per-parameter while maintaining reasonable perplexity and accuracy compared to scalar quantization. Our contributions include a novel compression technique that is compatible with state-of-the-art transformer quantization methods and can be seamlessly integrated into any hardware supporting 4-bit memory to reduce the model's memory footprint in memory-constrained devices. This work demonstrates a crucial step toward the efficient deployment of LLMs on edge devices.

CARVQ: Corrective Adaptor with Group Residual Vector Quantization for LLM Embedding Compression

Prior work has shown that a significant driver of performance in reasoning models is their ability to reason and self-correct. A distinctive marker in these reasoning traces is the token wait, which often signals reasoning behavior such as backtracking. Despite being such a complex behavior, little is understood of exactly why models do or do not decide to reason in this particular manner, which limits our understanding of what makes a reasoning model so effective. In this work, we address the question whether model's latents preceding wait tokens contain relevant information for modulating the subsequent reasoning process. To this end we train crosscoders at multiple layers layers of DeepSeek-R1-Distill-Llama-8B and its base version, and, introduce a novel latent attribution patching technique for the crosscoder setting. Using our technique, we locate a small set of features relevant for promoting/surpressing wait tokens' probabilities. Finally, through a targeted series of experiments analyzing max-activating examples and causal interventions, we show that many of our identified features indeed are relevant for the reasoning process and give rise to different types of reasoning patterns such as restarting from the beginning, recalling prior knowledge, expressing uncertainty, and double-checking.

Internal states before wait modulate reasoning patterns

LLM-as-a-judge evaluation metrics have gained popularity as an inexpensive and performant substitute for human evaluation. However, we find that the meta-evaluation setting in which the reliability of these LLM evaluators is established is substantially different from their use in model development. To address this, we propose a new meta-evaluation methodology that more closely aligns with practice by examining evaluators' ability to distinguish test system pairs that are closer in capability. Our fine-grained approach shows that LLM evaluator correlations with human judgments falls from ~0.8 to ~0.3 when evaluated in realistic settings, showcasing a key limitation of current norms. Equipped with this better methodology, we next analyze the impact that the choice of the reference model makes to LLM-as-a-judge evaluator performance. Our meta-evaluation strategy demonstrates that single-reference evaluators only perform well at ranking test systems that fall within particular capability ranges, even if the standard meta-evaluation reports high overall correlation. Taken together, our analysis shows critical issues with current LLM (meta-)evaluation and recommend avenues for improvement.

The Progress Illusion: Revisiting meta-evaluation standards of LLM evaluators

The escalating volume of academic research, coupled with a shortage of qualified reviewers, necessitates innovative approaches to peer review. In this work, we propose: (1) ReviewEval, a comprehensive evaluation framework for AI‐generated reviews that measures alignment with human assessments, verifies factual accuracy, assesses analytical depth, identifies degree of constructiveness and adherence to reviewer guidelines; and (2) ReviewAgent, an LLM‐based review generation agent featuring a novel alignment mechanism to tailor feedback to target conferences and journals, along with a self‐refinement loop that iteratively optimizes its intermediate outputs and an external improvement loop using ReviewEval to improve upon the final reviews. ReviewAgent improves actionable insights by 6.78% and 47.62% over existing AI baselines and expert reviews respectively. Further, it boosts analytical depth by 3.97% and 12.73%, enhances adherence to guidelines by 10.11% and 47.26% respectively. This paper establishes essential metrics for AI‐based peer review and substantially enhances the reliability and impact of AI‐generated reviews in academic research.

Downloads

Next from EMNLP 2025

Can Role Vectors Affect LLM Behaviour?

Similar lecture

Can Graph Descriptive Order Affect Solving Graph Problems with LLMs?