China

Current deep-research agents run in a &#39;&#39;fire-and-forget&#39;&#39; mode: once started, they give users no way to fix errors or add expert knowledge during execution. We present ResearStudio, the first open-source framework that places real-time human control at its core. The system follows a Collaborative Workshop design. A hierarchical Planner–Executor writes every step to a live &#39;&#39;plan-as-document,&#39;&#39; and a fast communication layer streams each action, file change, and tool call to a web interface. At any moment, the user can pause the run, edit the plan or code, run custom commands, and resume -- switching smoothly between AI-led, human-assisted and human-led, AI-assisted modes. In fully autonomous mode, ResearStudio achieves state-of-the-art results on the GAIA benchmark, surpassing systems like OpenAI&#39;s DeepResearch and Manus. These results show that strong automated performance and fine-grained human control can coexist. We will release the full code, protocol, and evaluation scripts to encourage further work on safe and controllable research agents upon acceptance.

EMNLP 2025

ResearStudio: A Human-intervenable Framework for Building Controllable Deep Research Agents

Current deep-research agents run in a ''fire-and-forget'' mode: once started, they give users no way to fix errors or add expert knowledge during execution. We present ResearStudio, the first open-source framework that places real-time human control at its core. The system follows a Collaborative Workshop design. A hierarchical Planner–Executor writes every step to a live ''plan-as-document,'' and a fast communication layer streams each action, file change, and tool call to a web interface. At any moment, the user can pause the run, edit the plan or code, run custom commands, and resume -- switching smoothly between AI-led, human-assisted and human-led, AI-assisted modes. In fully autonomous mode, ResearStudio achieves state-of-the-art results on the GAIA benchmark, surpassing systems like OpenAI's DeepResearch and Manus. These results show that strong automated performance and fine-grained human control can coexist. We will release the full code, protocol, and evaluation scripts to encourage further work on safe and controllable research agents upon acceptance.

demo

## 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."<br>

*Message from the General Chair, Dirk Hovy*

[**Link to Conference Handbook**](https://drive.google.com/file/d/1johU5QqVVYO4RfH7QcIORr7qrVBdzdwC/view?usp=sharing)





<br>

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

\emph{LLM-based Interactive Drama} introduces a novel dialogue scenario in which the player immerses into a character and engages in a dramatic story by interacting with characters played by LLM agents. Despite the fact that this emerging area holds significant promise, it remains largely underexplored due to the substantial resources required to develop a complete drama. This makes a significant barrier for researchers to replicate, extend, or study such systems. Hence, we present \textsc{Open-Theatre}, the first open-source toolkit for experiencing and developing LLM-based interactive drama. It integrates and advances multi-agent architectures and optimized prompts, allowing users to customize scripts and explore the LLM behavior in a lightweight manner. It also features a novel Retrieval-Augmented Generation (RAG)-based memory system, specifically designed to ensure character consistency and realistic long-term behavior in complex narratives. In addition, it provides a highly configurable pipeline, making it easy for researchers to optimize and develop and study new approaches.

Open-Theatre: An Open-Source Toolkit for LLM-based Interactive Drama

AutoIntent is an automated machine learning tool for text classification tasks. Unlike existing solutions, AutoIntent offers end-to-end automation with embedding model selection, classifier optimization, and decision threshold tuning, all within a modular, sklearn-like interface. The framework is designed to support multi-label classification and out-of-scope detection. AutoIntent demonstrates superior performance compared to existing AutoML tools on standard intent classification datasets and enables users to balance effectiveness and resource consumption.

AutoIntent: AutoML for Text Classification

Understanding the relationship between training data and model behavior during pretraining is crucial but existing workflows make this process cumbersome, fragmented, and often inaccessible to researchers. We present TokenSmith, an open-source library for interactive editing, inspection, and analysis of datasets used in Megatron-style pretraining frameworks such as GPT-NeoX, Megatron, and NVIDIA NeMo. TokenSmith supports a wide range of operations including searching, viewing, exporting, inspecting, and sampling data, all accessible through a simple user interface and a modular backend. It also enables structured editing of pretraining data without requiring changes to training code, simplifying dataset debugging, validation, and experimentation. TokenSmith is designed as a plug and play addition to existing large language model pretraining workflows, thereby democratizing access to production-grade dataset tooling. TokenSmith is hosted on GitHub (https://github.com/aflah02/tokensmith), with accompanying documentation and tutorials (https://aflah02.github.io/tokensmith/). A demonstration video is also available on YouTube (https://www.youtube.com/watch?v=cDO8VE9fZvU)

TokenSmith: Streamlining Data Editing, Search, and Inspection for Large-Scale Language Model Training and Interpretability

We introduce Co-DETECT (Collaborative Discovery of Edge cases in TExt ClassificaTion), a novel mixed-initiative annotation framework that integrates human expertise with automatic annotation guided by large language models (LLMs). Co-DETECT starts with an initial, sketch-level codebook and dataset provided by a domain expert, then leverages the LLM to annotate the data and identify edge cases that are not well described by the initial codebook. Specifically, Co-DETECT flags challenging examples, induces high-level, generalizable descriptions of edge cases, and assists user in incorporating edge case handling rules to improve the codebook. This iterative process enables more effective handling of nuanced phenomena through compact, generalizable annotation rules. Extensive user study, qualitative and quantitative analyses prove the effectiveness of Co-DETECT.

Co-DETECT: Collaborative Discovery of Edge Cases in Text Classification

The rapid advancement of Large Language Models (LLMs) in software engineering has revealed critical limitations in existing benchmarks, particularly the widely used SWE-bench dataset. Recent studies have uncovered severe data contamination issues, e.g. SWE-bench~\cite{jimenez2023swe} reports 32.67\% of successful patches involve direct solution leakage and 31.08\% pass due to inadequate test cases. We introduce \textbf{SWE-MERA}, a dynamic, continuously updated benchmark designed to address these fundamental challenges through an automated collection of real-world GitHub issues and rigorous quality validation. Our approach implements a reliable pipeline that ensures quality while minimizing contamination risks, resulting in approximately 10,000 potential tasks with 300 samples currently available. Evaluation using the Aider coding agent demonstrates strong discriminative power in state-of-the-art models. We report performance across dozen recent LLMs evaluated on tasks collected between September 2024 and June 2025.

SWE-MERA: A Dynamic Benchmark for Agenticly Evaluating Large Language Models on Software Engineering Tasks

Effective feedback is essential for student learning but is time-intensive for teachers. We present LearnLens, a modular, LLM-based system that generates personalised, curriculum-aligned feedback in science education. LearnLens comprises three components: (1) an error-aware assessment module that captures nuanced reasoning errors; (2) a curriculum-grounded generation module that uses a structured, topic-linked memory chain rather than traditional similarity-based retrieval, improving relevance and reducing noise; and (3) an educator-in-the-loop interface for customisation and oversight. LearnLens addresses key challenges in existing systems, offering scalable, high-quality feedback that empowers both teachers and students.

LearnLens: LLM-Enabled Personalised, Curriculum-Grounded Feedback with Educators in the Loop

We demonstrate Metamo, a browser-based dialogue system that transforms an off-the-shelf large language model into an empathetic coach for everyday workplace concerns. Metamo introduces a light, single-pass wrapper that first identifies the cognitive distortion behind an emotion, then recognizes the user’s emotion, and finally produces a question-centered reply that invites reflection, all within one model call. The wrapper keeps the response time below two seconds in the API, yet enriches the feedback with cognitively grounded insight. A front-end web interface renders the detected emotion as an animated avatar and shows distortion badges in real time, whereas a safety layer blocks medical advice and redirects crisis language to human hotlines. Empirical tests on public corpora confirmed that the proposed design improved emotion‑recognition quality and response diversity without sacrificing latency. A small user study with company staff reported higher perceived empathy and usability than a latency‑matched baseline. Metamo is model-agnostic, illustrating a practical path toward cognition-aware coaching tools.

Metamo: Empowering Large Language Models with Psychological Distortion Detection for Cognition-aware Coaching

Safety evaluation of Large Language Models (LLMs) has made progress and attracted academic interest, but it remains challenging to keep pace with the rapid integration of LLMs across diverse applications. Different applications expose users to various harms, necessitating application-specific safety evaluations with tailored harms and policies. Another major gap is the lack of focus on the dynamic and conversational nature of LLM systems. Such potential oversights can lead to harms that go unnoticed in standard safety benchmarks. This paper identifies the above as key requirements for robust LLM safety evaluation and recognizing that current evaluation methodologies do not satisfy these, we introduce the textttSAGE (Safety AI Generic Evaluation) framework. textttSAGE is an automated modular framework designed for customized and dynamic harm evaluations. It utilizes adversarial user models that are system-aware and have unique personalities, enabling a holistic red-teaming evaluation. We demonstrate textttSAGE's effectiveness by evaluating seven state-of-the-art LLMs across three applications and harm policies. Our experiments with multi-turn conversational evaluations revealed a concerning finding that harm steadily increases with conversation length. Furthermore, we observe significant disparities in model behavior when exposed to different user personalities and scenarios. Our findings also reveal that some models minimize harmful outputs by employing severe refusal tactics that can hinder their usefulness. These insights highlight the necessity of adaptive and context-specific testing to ensure better safety alignment and safer deployment of LLMs in real-world scenarios.

SAGE: A Generic Framework for LLM Safety Evaluation

This paper presents an audio chatbot system designed to handle a wide range of audio-related queries by integrating multiple specialized audio processing models. The proposed system uses an intent classifier, trained on a diverse audio query dataset, to route queries about audio content to expert models such as Automatic Speech Recognition (ASR), Speaker Diarization, Music Identification, and Text-to-Audio generation. A novel audio intent classification dataset is developed for building the intent classifier. A 3.8 B LLM model then takes inputs from an Audio Context Detection (ACD) module extracting audio event information from the audio and post processes text domain outputs from the expert models to compute the final response to the user. We evaluated the system on custom audio tasks and MMAU sound set benchmarks. The custom datasets were motivated by target use cases not covered in industry benchmarks. We proposed ACD-timestamp-QA (Question Answering) as well as ACD-temporal-QA datasets to evaluate timestamp and temporal reasoning questions, respectively. First, we determined that a BERT based Intent Classifier outperforms LLM-fewshot intent classifier in routing queries. Experiments further show that our approach significantly improves accuracy on some custom tasks compared to state-of-the-art Large Audio Language Models and outperforms models in the 7B parameter size range on the sound testset of the MMAU benchmark, thereby offering an attractive option for on device deployment.

Audio Query Handling System with Integrated Expert Models and Contextual Understanding

Large language models (LLMs) remain unreliable for global enterprise applications due to substantial performance gaps between high-resource and mid/low-resource languages, driven by English-centric pretraining and internal reasoning biases. This inconsistency undermines customer experience and operational reliability in multilingual settings such as customer support, content moderation, and information retrieval. Even with advanced Retrieval-Augmented Generation (RAG) systems, we observe up to an 29\% accuracy drop in non-English languages compared to English.

We propose a practical, batch-wise alignment strategy for fine-tuning LLMs, leveraging semantically equivalent multilingual data in each training batch to directly align model outputs across languages. This approach improves non-English accuracy by up to 23.9\% without compromising English performance, model reasoning, or retrieval quality. Our method is simple to implement, scalable, and integrates seamlessly with existing LLM training \& deployment pipelines, enabling more robust and equitable multilingual AI solutions in industry.

Downloads

Next from EMNLP 2025

Open-Theatre: An Open-Source Toolkit for LLM-based Interactive Drama

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from EMNLP 2025

Open-Theatre: An Open-Source Toolkit for LLM-based Interactive Drama

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads