ACL 2025

4,324 papers

Top Research Directions

Research directions most represented by ACL 2025 papers.

Factual Inaccuracy and Hallucination in Large Language Models 130
Shifting from Static Benchmarks to Dynamic Evaluation 120
Grounding Language Models with External Knowledge Retrieval 110
Evaluating Generative AI with Large Language Models 84
Intermediate Reasoning Chains for Deliberative Inference 80
Contextual Distraction and Noise Sensitivity in Language Models 78
Task Activation via In Context Demonstrations 66
Mitigating Data Scarcity in Low Resource Language Processing 65
Iterative Retrieval Augmented Generation for Multi Hop Reasoning 60
Autonomous Code Refinement via Execution Feedback Loops 60
The Language Bias of English Centric Benchmarks 54
Scaling Training Datasets Using Large Language Models 53
Understanding Prompt Sensitivity in Language Models 48
Role Specialization in Multi-Agent Architectures 47
Modeling Co Occurring and Conflicting Emotional States 47

Top Authors

AuthorPapers
Min Zhang26
Bing Qin19
Maosong Sun19
Minlie Huang17
Zhiyuan Liu16
Le Sun15
Fei Huang14
Preslav Nakov14
Huajun Chen13
Jiaheng Liu13
Xuanjing Huang13
Xuming Hu13
Arman Cohan12
Juntao Li12
Kang Liu12
Taro Watanabe12
Tat-Seng Chua12
Xianpei Han12
Xueqi Cheng12
Yangqiu Song12

Impactful Papers

Papers from ACL 2025 that shaped research clusters.

Foundational

Influential

Papers

Title
The Reasoning-Memorization Interplay in Language Models Is Mediated by a Single Direction
There's No Such Thing as Simple Reasoning for LLMs
The revision of linguistic annotation in the Universal Dependencies framework: a look at the annotators’ behavior
The Right Time Matters: Data Arrangement Affects Zero-Shot Generalization in Instruction Tuning
The Rise of Darkness: Safety-Utility Trade-Offs in Role-Playing Dialogue Agents
The Role of Abstract Representations and Observed Preferences in the Ordering of Binomials in Large Language Models
The Role of Exploration Modules in Small Language Models for Knowledge Graph Question Answering
The Role of Visual Modality in Multimodal Mathematical Reasoning: Challenges and Insights
The Rotary Position Embedding May Cause Dimension Inefficiency in Attention Heads for Long-Distance Retrieval
The Self-Improvement Paradox: Can Language Models Bootstrap Reasoning Capabilities without External Scaffolding?
The Silent Saboteur: Imperceptible Adversarial Attacks against Black-Box Retrieval-Augmented Generation Systems
The Structural Safety Generalization Problem
The Task Shield: Enforcing Task Alignment to Defend Against Indirect Prompt Injection in LLM Agents
The Threat of PROMPTS in Large Language Models: A System and User Prompt Perspective
The time scale of redundancy between prosody and linguistic context
The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs
The Tug of War Within: Mitigating the Fairness-Privacy Conflicts in Large Language Models
The Two Paradigms of LLM Detection: Authorship Attribution vs Authorship Verification
The UD-NewsCrawl Treebank: Reflections and Challenges from a Large-scale Tagalog Syntactic Annotation Project
The UNLP 2025 Shared Task on Detecting Social Media Manipulation
They want to pretend not to understand: The Limits of Current LLMs in Interpreting Implicit Content of Political Discourse
Think Again! The Effect of Test-Time Compute on Preferences, Opinions, and Beliefs of Large Language Models
Think Both Ways: Teacher-Student Bidirectional Reasoning Enhances MCQ Generation and Distractor Quality
Think&Cite: Improving Attributed Text Generation with Self-Guided Tree Search and Progress Reward Modeling
ThinkGuard: Deliberative Slow Thinking Leads to Cautious Guardrails
Thinking Before Running! Efficient Code Generation with Thorough Exploration and Optimal Refinement
Think Like a Person Before Responding: A Multi-Faceted Evaluation of Persona-Guided LLMs for Countering Hate Speech.
Think More, Hallucinate Less: Mitigating Hallucinations via Dual Process of Fast and Slow Thinking
THOR-MoE: Hierarchical Task-Guided and Context-Responsive Routing for Neural Machine Translation
TiC-LM: A Web-Scale Benchmark for Time-Continual LLM Pretraining
TicTac: Time-aware Supervised Fine-tuning for Automatic Text Dating
TIFIN India at SemEval-2025: Harnessing Translation to Overcome Multilingual IR Challenges in Fact-Checked Claim Retrieval
TigerLLM - A Family of Bangla Large Language Models
TILeN at SemEval-2025 Task 11: A Transformer-Based Model for Sentiment Classification Applied to the Russian Language
Time Course MechInterp: Analyzing the Evolution of Components and Knowledge in Large Language Models
Time-LlaMA: Adapting Large Language Models for Time Series Modeling via Dynamic Low-rank Adaptation
Time-MQA: Time Series Multi-Task Question Answering with Context Enhancement
Timestep Embeddings Trigger Collapse in Diffusion Text Generation
Time Travel: A Comprehensive Benchmark to Evaluate LMMs on Historical and Cultural Artifacts
tinaal at SemEval-2025 Task 11: Enhancing Perceived Emotion Intensity Prediction with Boosting Fine-Tuned Transformers
TituLLMs: A Family of Bangla LLMs with Comprehensive Benchmarking
TLPIQ at BioLaySumm: Hide and Seq, a FLAN-T5 Model for Biomedical Summarization
TN-Eval: Rubric and Evaluation Protocols for Measuring the Quality of Behavioral Therapy Notes
To Code or not to Code? Adaptive Tool Integration for Math Language Models via Expectation-Maximization
TokAlign: Efficient Vocabulary Adaptation via Token Alignment
Token-Budget-Aware LLM Reasoning
Tokenisation is NP-Complete
Tokenization is Sensitive to Language Variation
Token-level Preference Self-Alignment Optimization for Multi-style Outline Controllable Generation
Token-level semantic typology without a massively parallel corpus

« prev Page 77/87 next »