China

Large language models (LLMs) have emerged as powerful tools for developing educational systems. While previous studies have explored modeling student mistakes, a critical gap remains in understanding whether LLMs can generate correct solutions that represent student responses to free-response problems. In this paper, we compare the distribution of solutions produced by LLMs with various sampling and prompting techniques and those generated by students, using conversations where students teach math problems to a conversational robot. Our study reveals discrepancies between the correct solutions produced by LLMs and by students. We discuss the practical implications of these findings for the design and evaluation of LLM-supported educational systems.

EMNLP 2025

Can LLMs simulate the same correct solutions to free-response math problems as real students?

math

educational application

## Welcome to EMNLP 2024! 
We are excited to welcome you to one of the most prominent conferences in the field of Natural Language Processing. This year, EMNLP 2024 is being held in a hybrid format,
offering both virtual and in-person participation in beautiful Miami. Due to a record-breaking number of submissions, we've expanded the total number of accepted papers to accommodate more cutting-edge research from around the globe.
### [Conference Handbook](https://drive.google.com/file/d/1WPROgxjLAC96AJL7Ugy0tEnYm7dkrbHt/view?usp=sharing)

You are required to register for this event. **Please register [here](https://2024.emnlp.org/registration/).** The EMNLP 2024 event page on Underline will be open to public one week prior to the event.

Please register!

EMNLP 2024

EMNLP 2024 will take place in Miami, Florida from Nov 12th to Nov 16th, 2024, at the Hyatt Regency Miami Hote and on Underline for remote participants.

technical paper

## 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.

Automating Exploratory Data Analysis (EDA) is critical for accelerating the workflow of data scientists. While Large Language Models (LLMs) offer a promising solution, current LLM-only approaches often exhibit limited accuracy and code reliability on less-studied or private datasets. Moreover, their effectiveness significantly diminishes with open-source LLMs compared to proprietary ones, limiting their usability in enterprises that prefer local models for privacy and cost. To address these limitations, we introduce RAGvis, a novel two-stage Retrieval-Augmented Generation (RAG) framework. RAGvis first builds upon and enriches a base knowledge graph of EDA notebooks with structured EDA operation semantics. These semantics are extracted by an LLM guided by our empirically-developed EDA operations taxonomy. Subsequently, during online generation for new datasets, RAGvis retrieves relevant operations, aligns them to the dataset's structure, refines them using LLM reasoning, and employs a self-correcting agent to generate executable Python code. Experiments on two benchmarks demonstrate that RAGvis significantly improves code executability (pass rate) and semantic accuracy in generated operations. This enhanced performance is achieved with substantially lower token usage compared to LLM-only baselines. Notably, our approach enables smaller, open-source LLMs to match the performance of proprietary models, presenting a reliable and cost-effective pathway for automated EDA code generation.

Reliable and Cost-Effective Exploratory Data Analysis via Graph-Guided RAG

Recently, Large Language Models (LLMs) have been widely adopted in a wide range of tasks, leading to increasing attention towards the research on how scaling LLMs affects their performance. Existing works, termed Scaling Laws, have discovered that the final test loss of LLMs scales as power-laws with model size, computational budget, and dataset size. However, the temporal change of the test loss of an LLM throughout its pretraining process remains unexplored, though it is valuable in many aspects, such as selecting better hyperparameters \textit{directly} on the target LLM. In this paper, we propose the novel concept of Temporal Scaling Law, studying how the test loss of an LLM evolves as the training steps scale up. In contrast to modeling the test loss as a whole in a coarse-grained manner, we break it down and dive into the fine-grained test loss of each token position, and further develop a dynamic hyperbolic-law. Afterwards, we derive the much more precise temporal scaling law by studying the temporal patterns of the parameters in the dynamic hyperbolic-law. Results on both in-distribution (ID) and out-of-distribution (OOD) validation datasets demonstrate that our temporal scaling law accurately predicts the test loss of LLMs across training steps. Our temporal scaling law has broad practical applications. First, it enables direct and efficient hyperparameter selection on the target LLM, such as data mixture proportions. Secondly, viewing the LLM pretraining dynamics from the token position granularity provides some insights to enhance the understanding of LLM pretraining.

Temporal Scaling Law for Large Language Models

The recent rise of reasoning-tuned Large Language Models (LLMs)—which generate chains of thought (CoTs) before giving the final answer—has attracted significant attention and offers new opportunities for gaining insights into human label variation, which refers to plausible differences in how multiple annotators label the same data instance. Prior work has shown that LLM-generated explanations can help align model predictions with human label distributions, but typically adopt a *reverse* paradigm: producing explanations based on given answers. In contrast, CoTs provide a *forward* reasoning path that may implicitly embed rationales for each answer option, before generating the answers. We thus propose a novel LLM-based pipeline enriched with linguistically-grounded discourse segmenters to extract supporting and opposing statements for each answer option from CoTs with improved accuracy. We also propose a rank-based HLV evaluation framework that prioritizes the ranking of answers over exact scores, which instead favor direct comparison of label distributions. Our method outperforms a direct generation method as well as baselines on three datasets, and shows better alignment of ranking methods with humans, highlighting the effectiveness of our approach.

Threading the Needle: Reweaving Chain-of-Thought Reasoning to Explain Human Label Variation

In large-scale industrial LLM systems, prompt templates often expand to thousands of tokens as teams iteratively incorporate sections such as task instructions, few-shot examples, and heuristic rules to enhance robustness and coverage. This expansion leads to bloated prompts that are difficult to maintain and incur significant inference latency and serving costs. To address this, we introduce Prompt Compression via Attribution Estimation (ProCut), a flexible, LLM-agnostic, training-free framework that compresses prompts through attribution analysis. ProCut segments prompt templates into semantically meaningful units, quantifies their impact on task performance, and prunes low-utility components. Through extensive experiments on five public benchmark datasets and real-world industrial prompts, we show that ProCut achieves substantial prompt size reductions (78% fewer tokens in production) while maintaining or even slightly improving task performance (up to 62% better than alternative methods). We further introduce an LLM-driven attribution estimator that reduces compression latency by over 50%, and demonstrate that ProCut integrates seamlessly with existing prompt-optimization frameworks to produce concise, high-performing prompts.

ProCut: LLM Prompt Compression via Attribution Estimation

Child-Directed Speech (CDS) holds unique linguistic properties that distinguish it from other types of textual corpora. Language models trained using CDS often obtain superior results compared with the same size of different types of data. Several studies have aimed at modifying non-CDS data to mimic its linguistic properties to match the hypothesized advantageous aspects of CDS. Here, we propose to adapt the non-CDS portions of the training data to include questions similar to CDS interaction. We modify the data by adding artificially generated questions to the data and methodically analyzing the change in performance using each modified dataset. Our results show that artificial question generation strongly depends on the properties of the original dataset. While the performance improves for question-related measures, the overall performance is negatively affected as a result of the reduced syntactic diversity.

What did you say? Generating Child-Directed Speech Questions to Train LLMs

Large Language Models (LLMs) have achieved impressive performance in text summarization and are increasingly deployed in real-world applications. However, these systems often inherit associative and framing biases from pre-training data, leading to inappropriate or unfair outputs in downstream tasks. In this work, we present AdvSumm (Adversarial Summarization), a domain-agnostic training framework designed to mitigate bias in text summarization through improved generalization. Inspired by adversarial robustness, AdvSumm introduces a novel Perturber component that applies gradient-guided perturbations at the embedding level of Sequence-to-Sequence models, enhancing the model’s robustness to input variations. We empirically demonstrate that AdvSumm effectively reduces different types of bias in summarization—specifically, name-nationality bias and political framing bias—without compromising summarization quality. Compared to standard transformers and data augmentation techniques like back-translation, AdvSumm achieves stronger bias mitigation performance across benchmark datasets.

AdvSumm: Adversarial Training for Bias Mitigation in Text Summarization

Black-box verifiers for abstractive summaries often struggle with complex claims that require multi-hop reasoning, and they typically provide a single verdict without an interpretable rationale. As a result, it becomes difficult to understand or audit their failures. We address this with HalluTree, a framework that models verification as an interpretable claim tree. HalluTree first decomposes summaries into subclaims, classifying each into two types -- extractive (directly verifiable against evidence) or inferential (requiring reasoning) -- which follow distinct verification paths. Extractive claims are robustly verified against evidence using an ensemble of lightweight NLI models. Crucially, inferential claims trigger a process that generates a natural program -- an explicit reasoning chain that integrates supporting evidence and logical steps -- which is then executed to determine the claim's validity. Evaluation on the LLM-AggreFact benchmark demonstrates HalluTree's effectiveness: it achieves performance competitive with top-tier black-box models, including Bespoke-MiniCheck, while providing transparent and auditable reasoning programs for every inferential judgment. This combination of competitive accuracy and high interpretability offers a significant advance over opaque, single-classification verifiers. We will publically release code, data, prompts, and other artifacts upon acceptance.

HalluTree: Explainable Multi-Hop Hallucination Detection for Abstractive Summarization

Topic models represent topics as ranked term lists, which are often hard to interpret in scientific domains. We explore Topic Description for Scientific Corpora, an approach to generating structured summaries for topic-specific document sets. We propose and investigate two LLM-based pipelines: Selective Context Summarisation (SCS), which uses maximum marginal relevance to select representative documents; and Compressed Context Summarisation (CCS), a hierarchical approach that compresses document sets through iterative summarisation. We evaluate both methods using SUPERT and multi-model LLM-as-a-Judge across three topic modeling backbones and three scientific corpora. Our preliminary results suggest that SCS tends to outperform CCS in quality and robustness, while CCS shows potential advantages on larger topics. Our findings highlight interesting trade-offs between selective and compressed strategies for topic-level summarisation in scientific domains. We release code and data for two of the three datasets.

From Keyterms to Context: Exploring Topic Description Generation in Scientific Corpora

Large language models (LLMs) have achieved impressive performance in text summarization, yet their performance often falls short when applied to specialized domains that differ from their original pre-training distribution. While fine-tuning can improve summarization quality, it typically relies on costly and scarce high-quality labeled data. In this work, we explore continual pre-training as a scalable, self-supervised approach to adapt LLMs for downstream summarization tasks, particularly in the context of noisy real-world conversation transcripts. We conduct extensive experiments using large-scale, unlabeled business conversation data to investigate whether continual pre-training enhances model capabilities in conversational summarization. Our results demonstrate that continual pre-training yields substantial gains in both in-domain and out-of-domain summarization benchmarks, while maintaining strong generalization and robustness. We also analyze the effects of data selection strategies, providing practical guidelines for applying continual pre-training in summarization-focused industrial applications.

DACP: Domain-Adaptive Continual Pre-Training of Large Language Models for Phone Conversation Summarization

Individuals express diverse opinions, a fair summary should represent these viewpoints comprehensively.
Previous research on fairness in opinion summarisation using large language models (LLMs) relied on hyperparameter tuning or providing ground truth distributional information in prompts. However, these methods face practical limitations: end-users rarely modify default model parameters, and accurate distributional information is often unavailable. Building upon cognitive science research demonstrating that frequency-based representations reduce systematic biases in human statistical reasoning by making reference classes explicit and reducing cognitive load, this study investigates whether frequency framed prompting (REFER) can similarly enhance fairness in LLM opinion summarisation. Through systematic experimentation with different prompting frameworks, we adapted techniques known to improve human reasoning to elicit more effective information processing in language models compared to abstract probabilistic representations. Our results demonstrate that REFER enhances fairness in language models when summarising opinions. 
This effect is particularly pronounced in larger language models and using stronger reasoning instructions.

Downloads

Next from EMNLP 2025

Reliable and Cost-Effective Exploratory Data Analysis via Graph-Guided RAG

Similar lecture

Adversarial Math Word Problem Generation

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES