China

Recently, Test-Time Scaling (TTS) has gained increasing attention for improving LLM reasoning performance at test time without retraining the model. A notable TTS technique is Self-Consistency (SC), which generates multiple reasoning chains in parallel and selects the final answer via majority voting. While effective, the order-of-magnitude computational overhead limits its broad deployment. Prior attempts to accelerate SC mainly rely on model-based confidence scores or heuristics with limited empirical support. For the first time, we theoretically and empirically analyze the inefficiencies of SC and reveal actionable opportunities for improvement. Building on these insights, we propose Slim-SC, a (thinking) step-wise pruning strategy that removes redundant chains using inter-chain similarity at the thought level. Experiments on three mathematical reasoning datasets and two recent LLM architectures confirm that Slim-SC not only reduces resource waste but also matches or even increases the accuracy of SC, providing a simple yet effective alternative for efficient test-time scaling with Self-Consistency.

EMNLP 2025

Slim-SC: Thought Pruning for Efficient Scaling with Self-Consistency

test-time scaling

efficient reasoning

self-consistency

### Welcome!
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. Our Virtual Poster Sessions will take place online Thursday, August 22, 2024.

You are required to register for this event. **Please register [here](https://2024.aclweb.org/registration). **

If you have already registered, please check your inbox for an email from Underline granting you access to ACL 2024 content.

Please register!

ACL 2024

The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. More information will be announced soon.

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.

Reinforcement learning from human feedback (RLHF) and, at its core, reward modeling have become a crucial part of training powerful large language models (LLMs). One commonly overlooked factor in training high-quality reward models (RMs) is the effect of the base model, which is becoming more challenging to choose given the rapidly growing pool of LLMs. In this work, we present a systematic analysis of the effect of base model selection on reward modeling performance. Our results show that the performance can be improved by up to 14% compared to the most common (i.e., default) choice. Moreover, we showcase the strong statistical relation between some existing benchmarks and downstream performances. We also demonstrate that the results from a small set of benchmarks could be combined to boost the model selection (+18% on average in the top 5-10). Lastly, we illustrate the impact of different post-training steps on the final performance and explore using estimated data distributions to reduce performance prediction error.

A Systematic Analysis of Base Model Choice for Reward Modeling

Detecting self-contradictions within documents is a challenging task for ensuring textual coherence and reliability. While large language models (LLMs) have advanced in many natural language understanding tasks, document-level self-contradiction detection (DSCD) remains insufficiently studied. Recent approaches leveraging Chain-of-Thought (CoT) prompting aim to enhance reasoning and interpretability; however, they only gain marginal improvement and often introduce inconsistencies across repeated responses. We observe that such inconsistency arises from incomplete reasoning chains that fail to include all relevant contradictory sentences consistently. To address this, we propose a two-stage method that combines supervised fine-tuning (SFT) and reinforcement learning (RL) to enhance DSCD performance. In the SFT phase, a teacher model helps the model learn reasoning patterns, while RL further refines its reasoning ability. Our method incorporates a task-specific reward function to expand the model's reasoning scope, boosting both accuracy and consistency. On the ContraDoc benchmark, our approach significantly boosts Llama 3.1-8B-Instruct's accuracy from 38.5% to 51.1%, and consistency from 59.6% to76.2%.

Think Wider, Detect Sharper: Reinforced Reference Coverage for Document-Level Self-Contradiction Detection

Although prior work on bias mitigation has focused on promoting social equality and demographic parity, less attention has been given to aligning language model outputs with desired distributions. For example, we might want to align a model with real-world distributions to support factual grounding. Thus, we define bias as deviation from a desired distribution, which may be an equal or real-world distribution — depending on application goals. We propose a weighted adaptive loss based fine-tuning method that aligns LLM’s gender–profession output distribution with the desired distribution, while preserving language modeling performance. Using three profession sets—male-dominated, female-dominated, and gender-balanced— derived from U.S. labor statistics (2024), we evaluate both our adaptive method for reflecting reality and a non-adaptive variant for equality. Experiments on three masked language models and one autoregressive model (LLaMA 3.2-3B-Instruct) show near-complete bias mitigation under equality and about 30–75% (MLMs) and 50% (ALM) bias reduction, when aligning LLMs to distributions in the real world.

LLM Bias Detection and Mitigation through the Lens of Desired Distributions

The Mixture of Depths (MoD) was introduced to improve computational efficiency by dynamically skipping less important layers, reducing redundant computation while maintaining model capacity. Despite its promise, existing MoD approaches remain under-explored and face two main challenges: (1) \textit{high training costs due to the need to train the entire model along with the routers that determine which layers to skip}, and (2) \textit{performance degradation when important layers are bypassed}. In response to the first issue, we propose Router-Tuning, which fine-tunes only the routers on a small dataset, drastically reducing the computational overhead associated with full model training. For the second challenge, we investigate \method~across different architectures and granularities, demonstrating its effectiveness on Attention layers and MoE layers. This method preserves the model’s performance while significantly enhancing computational and memory efficiency. Extensive experiments demonstrate that our approach delivers competitive results while dramatically improving the computation efficiency, e.g., 21% speedup and only a 0.2% performance drop. The code will be released upon acceptance.

Router-Tuning: A Simple and Effective Approach for Dynamic Depth

Advancements in Large Language Models (LLMs) and their increasing use in medical question-answering necessitate rigorous evaluation of their reliability. A critical challenge lies in hallucination, where models generate plausible yet factually incorrect outputs. In the medical domain, this poses serious risks to patient safety and clinical decision-making. To address this, we introduce, the first benchmark specifically designed for medical hallucination detection. MedHallu comprises 10,000 high-quality question-answer pairs derived from PubMedQA, with hallucinated answers systematically generated through a controlled pipeline. Our experiments show that state-of-the-art LLMs, including GPT-4o, Llama-3.1, and the medically fine-tuned UltraMedical, struggle with this binary hallucination detection task, with the best model achieving an F1 score as low as 0.625 for detecting ''hard'' category hallucinations. Using bidirectional entailment clustering, we show that harder-to-detect hallucinations are semantically closer to ground truth. Through experiments, we also show incorporating domain-specific knowledge and introducing a ''not sure'' category as one of the answer categories improves the precision and F1 scores by up to 38% relative to baselines.

MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models

Rencent advancements in large language models (LLM) have shown impressive versatility across various tasks. Short text matching is one of the fundamental technologies in natural language processing. In previous studies, the common approach to applying them to Chinese is segmenting each sentence into words, and then taking these words as input. However, existing approaches have three limitations: 1) Some Chinese words are polysemous, and semantic information is not fully utilized. 2) Some models suffer potential issues caused by word segmentation and incorrect recognition of negative words affects the semantic understanding of the whole sentence. 3) Fuzzy negation words in ancient Chinese are difficult to recognize and match. In this work, we propose a novel adaptive Transformer for Chinese short text matching using Data Augmentation and Semantic Awareness (DASA), which can fully mine the information expressed in Chinese text to deal with word ambiguity. DASA is based on a Graph Attention Transformer Encoder that takes two word lattice graphs as input and integrates sense information from N-HowNet to moderate word ambiguity. Specially, we use an LLM to generate similar sentences for the optimal text representation. Experimental results show that the augmentation done using DASA can considerably boost the performance of our system and achieve significantly better results than previous state-of-the-art methods on four available datasets, namely MNS, LCQMC, AFQMC, and BQ.

DASA-Trans-STM: Adaptive Efficient Transformer for Short Text Matching using Data Augmentation and Semantic Awareness

Large language models (LLMs) have shown strong potential in complex reasoning tasks. However, as task complexity increases, their performance often degrades, resulting in hallucinations, errors, and logical inconsistencies. To enhance reasoning capabilities, Monte Carlo Tree Search (MCTS) has been introduced to guide the exploration of reasoning paths in a structured manner. Despite its advantages, traditional MCTS relies on fixed reasoning strategies, limiting the diversity of reasoning paths and the coverage of the solution space. To address these limitations, we propose Dynamic Strategy-Guided MCTS (DSG-MCTS), a novel framework that dynamically integrates multiple reasoning strategies, such as abductive and analogical reasoning, to expand the reasoning space. At the same time, DSG-MCTS enhances reasoning efficiency through a dynamic strategy selection mechanism that adapts to the task context. Experimental results on challenging reasoning benchmarks demonstrate that DSG-MCTS achieves improved accuracy and efficiency, outperforming existing state-of-the-art methods. The implementation is available at: https://anonymous.4open.science/r/DSG-MCTS-27AB.

DSG-MCTS: A Dynamic Strategy-Guided Monte Carlo Tree Search for Diversified Reasoning in Large Language Models

Social text understanding is prone to fail when opinions are conveyed implicitly or sarcastically. It is therefore desired to model users' contexts in processing the texts authored by them. In this work, we represent users within a social embedding space that was learned from Twitter at large-scale. Similar to word embeddings that encode lexical semantics, the user embeddings encode general dimensions of social semantics. We conduct comprehensive experiments of author-informed stance prediction, showing improved learning generalization due to user modeling both within and across topics. Promising results are further shown for author-informed toxicity detection. The proposed approach may pave way to social NLP that considers user embeddings as contextual modality. Investigating the social factors that underlie author modeling, we find however that user stances as labeled in benchmark datasets exhibit socio-demographic patterns. Hence, author-informed stance detection approaches can both utilize, and should be aware of social biases.

 Towards Author-informed NLP: Mind the Social Bias

Large language models segment many words into multiple tokens, and companies that make those models claim that meaningful subword tokens are essential. To investigate whether subword tokens bear meaning, we segmented tens of thousands of words from each of 41 languages according to three generations of GPT tokenizers. We found that words which share tokens are more semantically similar than expected by chance or expected from length alone, that tokens capture morphological information even when they don’t look like morphemes, and that tokens capture more information than is explained by morphology. In languages that use a script other than the Latin alphabet, GPT-4 tokens are uninformative, but GPT-4o has improved this situation. These results suggest that comparing tokens to morphemes overlooks the wider variety of semantic information available in word form and that standard tokenization methods successfully capture much of that information.

How much Semantic Information is Available in Large Language Model Tokens

Pre-trained language models (PLMs) have driven substantial progress in natural language processing but remain vulnerable to adversarial attacks, raising concerns about their robustness in real-world applications. Previous studies have sought to mitigate the impact of adversarial attacks by introducing adversarial perturbations into the training process, either implicitly or explicitly. While both strategies enhance robustness, they often incur high computational costs. In this work, we propose a simple yet effective add-on module that enhances the adversarial robustness of PLMs by removing instance-level principal components, without relying on conventional adversarial defences or perturbing the original training data. Our approach transforms the embedding space to approximate Gaussian properties, thereby reducing its susceptibility to adversarial perturbations while preserving semantic relationships. This transformation aligns embedding distributions in a way that minimises the impact of adversarial noise on decision boundaries, enhancing robustness without requiring adversarial examples or costly training-time augmentation. Evaluations on eight benchmark datasets show that our approach improves adversarial robustness while maintaining comparable before-attack accuracy to baselines, achieving a balanced trade-off between robustness and generalisation.

Downloads

Next from EMNLP 2025

A Systematic Analysis of Base Model Choice for Reward Modeling

Similar lecture

Soft Self-Consistency Improves Language Models Agents

Stay up to date with the latest Underline news!

PRODUCT

COMPANY

RESOURCES