Natural Language Processing Techniques

Neueste Paper

• Why Do Multilingual Reasoning Gaps Emerge in Reasoning Language Models?

  Deokhyung Kang, Seonjeong Hwang, Daehui Kim, Hyounghun Kim, Gary Geunbae Lee · 3. November 2025

  Reasoning language models (RLMs) achieve strong performance on complex reasoning tasks, yet they still suffer from a multilingual reasoning gap, performing better in high-resource languages than in low-resource ones. While recent efforts have reduced this gap, its underlying causes remain largely un…

• (How) Do Language Models Track State?

  Belinda Z. Li, Zifan Carl Guo, Jacob Andreas · 3. November 2025

  Transformer language models (LMs) exhibit behaviors -- from storytelling to code generation -- that seem to require tracking the unobserved state of an evolving world. How do they do this? We study state tracking in LMs trained or fine-tuned to compose permutations (i.e., to compute the order of a s…

• Language Model Preference Evaluation with Multiple Weak Evaluators

  Zhengyu Hu, Jieyu Zhang, Zhihan Xiong, Alexander Ratner, Kaize Ding, Ranjay Krishna · 31. Oktober 2025

  Despite the remarkable success of Large Language Models (LLMs), evaluating their outputs' quality regarding preference remains a critical challenge. While existing works usually leverage a strong LLM as the judge for comparing LLMs' response pairwisely, such a single-evaluator approach is vulnerable…

• Scales++: Compute Efficient Evaluation Subset Selection with Cognitive Scales Embeddings

  Andrew M. Bean, Nabeel Seedat, Shengzhuang Chen, Jonathan Richard Schwarz · 31. Oktober 2025

  The prohibitive cost of evaluating large language models (LLMs) on comprehensive benchmarks necessitates the creation of small yet representative data subsets (i.e., tiny benchmarks) that enable efficient assessment while retaining predictive fidelity. Current methods for this task operate under a m…

• Revisiting Multilingual Data Mixtures in Language Model Pretraining

  Negar Foroutan, Paul Teiletche, Ayush Kumar Tarun, Antoine Bosselut · 31. Oktober 2025

  The impact of different multilingual data mixtures in pretraining large language models (LLMs) has been a topic of ongoing debate, often raising concerns about potential trade-offs between language coverage and model performance (i.e., the curse of multilinguality). In this work, we investigate thes…

• Inference-Cost-Aware Dynamic Tree Construction for Efficient Inference in Large Language Models

  Yinrong Hong, Zhiquan Tan, Kai Hu · 31. Oktober 2025

  Large Language Models (LLMs) face significant inference latency challenges stemming from their autoregressive design and large size. To address this, speculative decoding emerges as a solution, enabling the simultaneous generation and validation of multiple tokens. While recent approaches like EAGLE…

• Unravelling the Mechanisms of Manipulating Numbers in Language Models

  Michal \v{S}tef\'anik, Timothee Mickus, Marek Kadl\v{c}\'ik, Bertram H{\o}jer, Michal Spiegel, Ra\'ul V\'azquez, Aman Sinha, Josef Kucha\v{r}, Philipp Mondorf · 31. Oktober 2025

  Recent work has shown that different large language models (LLMs) converge to similar and accurate input embedding representations for numbers. These findings conflict with the documented propensity of LLMs to produce erroneous outputs when dealing with numeric information. In this work, we aim to e…

• Diversity as a Reward: Fine-Tuning LLMs on a Mixture of Domain-Undetermined Data

  Zhenqing Ling, Daoyuan Chen, Liuyi Yao, Qianli Shen, Yaliang Li, Ying Shen · 31. Oktober 2025

  Fine-tuning large language models (LLMs) using diverse datasets is crucial for enhancing their overall performance across various domains. In practical scenarios, existing methods based on modeling the mixture proportions of data composition often struggle with data whose domain labels are missing, …

• PORTool: Tool-Use LLM Training with Rewarded Tree

  Feijie Wu, Weiwu Zhu, Yuxiang Zhang, Soumya Chatterjee, Jiarong Zhu, Fan Mo, Rodin Luo, Jing Gao · 31. Oktober 2025

  Current tool-use large language models (LLMs) are trained on static datasets, enabling them to interact with external tools and perform multi-step, tool-integrated reasoning, which produces tool-call trajectories. However, these models imitate how a query is resolved in a generic tool-call routine, …

• Polybasic Speculative Decoding Through a Theoretical Perspective

  Ruilin Wang, Huixia Li, Yuexiao Ma, Xiawu Zheng, Fei Chao, Xuefeng Xiao, Rongrong Ji · 31. Oktober 2025

  Inference latency stands as a critical bottleneck in the large-scale deployment of Large Language Models (LLMs). Speculative decoding methods have recently shown promise in accelerating inference without compromising the output distribution. However, existing work typically relies on a dualistic dra…

• Rethinking Optimal Verification Granularity for Compute-Efficient Test-Time Scaling

  Hao Mark Chen, Guanxi Lu, Yasuyuki Okoshi, Zhiwen Mo, Masato Motomura, Hongxiang Fan · 31. Oktober 2025

  Test-time scaling (TTS) has proven effective in enhancing the reasoning capabilities of large language models (LLMs). Verification plays a key role in TTS, simultaneously influencing (1) reasoning performance and (2) compute efficiency, due to the quality and computational cost of verification. In t…

• ReSpec: Towards Optimizing Speculative Decoding in Reinforcement Learning Systems

  Qiaoling Chen, Zijun Liu, Peng Sun, Shenggui Li, Guoteng Wang, Ziming Liu, Yonggang Wen, Siyuan Feng, Tianwei Zhang · 31. Oktober 2025

  Adapting large language models (LLMs) via reinforcement learning (RL) is often bottlenecked by the generation stage, which can consume over 75\% of the training time. Speculative decoding (SD) accelerates autoregressive generation in serving systems, but its behavior under RL training remains largel…

• PRESTO: Preimage-Informed Instruction Optimization for Prompting Black-Box LLMs

  Jaewon Chu, Seunghun Lee, Hyunwoo J. Kim · 31. Oktober 2025

  Large language models (LLMs) have achieved remarkable success across diverse domains, due to their strong instruction-following capabilities. This has led to increasing interest in optimizing instructions for black-box LLMs, whose internal parameters are inaccessible but widely used due to their str…

• The Structure of Relation Decoding Linear Operators in Large Language Models

  Miranda Anna Christ, Adri\'an Csisz\'arik, Gergely Becs\'o, D\'aniel Varga · 31. Oktober 2025

  This paper investigates the structure of linear operators introduced in Hernandez et al. [2023] that decode specific relational facts in transformer language models. We extend their single-relation findings to a collection of relations and systematically chart their organization. We show that such c…

• GnnXemplar: Exemplars to Explanations -- Natural Language Rules for Global GNN Interpretability

  Burouj Armgaan, Eshan Jain, Harsh Pandey, Mahesh Chandran, Sayan Ranu · 30. Oktober 2025

  Graph Neural Networks (GNNs) are widely used for node classification, yet their opaque decision-making limits trust and adoption. While local explanations offer insights into individual predictions, global explanation methods, those that characterize an entire class, remain underdeveloped. Existing …

• Doubly Robust Alignment for Large Language Models

  Erhan Xu, Kai Ye, Hongyi Zhou, Luhan Zhu, Francesco Quinzan, Chengchun Shi · 30. Oktober 2025

  This paper studies reinforcement learning from human feedback (RLHF) for aligning large language models with human preferences. While RLHF has demonstrated promising results, many algorithms are highly sensitive to misspecifications in the underlying preference model (e.g., the Bradley-Terry model),…

• Sequences of Logits Reveal the Low Rank Structure of Language Models

  Noah Golowich, Allen Liu, Abhishek Shetty · 30. Oktober 2025

  A major problem in the study of large language models is to understand their inherent low-dimensional structure. We introduce an approach to study the low-dimensional structure of language models at a model-agnostic level: as sequential probabilistic models. We first empirically demonstrate that a w…

• Non-Markovian Discrete Diffusion with Causal Language Models

  Yangtian Zhang, Sizhuang He, Daniel Levine, Lawrence Zhao, David Zhang, Syed A Rizvi, Shiyang Zhang, Emanuele Zappala, Rex Ying, David van Dijk · 30. Oktober 2025

  Discrete diffusion models offer a flexible, controllable approach to structured sequence generation, yet they still lag behind causal language models in expressive power. A key limitation lies in their reliance on the Markovian assumption, which restricts each step to condition only on the current s…

• SATURN: SAT-based Reinforcement Learning to Unleash Language Model Reasoning

  Huanyu Liu, Jia Li, Hao Zhu, Kechi Zhang, Yihong Dong, Ge Li · 30. Oktober 2025

  How to design reinforcement learning (RL) tasks that effectively unleash the reasoning capability of large language models (LLMs) remains an open question. Existing RL tasks (e.g., math, programming, and constructing reasoning tasks) suffer from three key limitations: (1) Scalability. They rely heav…

• Beyond Neural Incompatibility: Easing Cross-Scale Knowledge Transfer in Large Language Models through Latent Semantic Alignment

  Jian Gu, Aldeida Aleti, Chunyang Chen, Hongyu Zhang · 29. Oktober 2025

  Large Language Models (LLMs) encode vast amounts of knowledge in their massive parameters, which is accessible to locate, trace, and analyze. Despite advances in neural interpretability, it is still not clear how to transfer knowledge in a fine-grained manner, namely parametric knowledge transfer (P…

• Zero-Shot Cross-Lingual Transfer using Prefix-Based Adaptation

  Snegha A (Indian Institute of Technology Bombay), Sayambhu Sen (Amazon Alexa), Piyush Singh Pasi (Amazon Alexa), Abhishek Singhania (Amazon Alexa), Preethi Jyothi (Indian Institute of Technology Bombay) · 29. Oktober 2025

  With the release of new large language models (LLMs) like Llama and Mistral, zero-shot cross-lingual transfer has become increasingly feasible due to their multilingual pretraining and strong generalization capabilities. However, adapting these decoder-only LLMs to new tasks across languages remains…

• One-Step is Enough: Sparse Autoencoders for Text-to-Image Diffusion Models

  Viacheslav Surkov, Chris Wendler, Antonio Mari, Mikhail Terekhov, Justin Deschenaux, Robert West, Caglar Gulcehre, David Bau · 29. Oktober 2025

  For large language models (LLMs), sparse autoencoders (SAEs) have been shown to decompose intermediate representations that often are not interpretable directly into sparse sums of interpretable features, facilitating better control and subsequent analysis. However, similar analyses and approaches h…

• PTQTP: Post-Training Quantization to Trit-Planes for Large Language Models

  He Xiao, Runming Yang, Qingyao Yang, Wendong Xu, Zhen Li, Yupeng Su, Zhengwu Liu, Hongxia Yang, Ngai Wong · 29. Oktober 2025

  Post-training quantization (PTQ) of large language models (LLMs) to extremely low bit-widths remains challenging due to the fundamental trade-off between computational efficiency and model expressiveness. While existing ultra-low-bit PTQ methods rely on binary approximations or complex compensation …

• Think Just Enough: Sequence-Level Entropy as a Confidence Signal for LLM Reasoning

  Aman Sharma, Paras Chopra · 29. Oktober 2025

  We introduce a simple, yet novel entropy-based framework to drive token efficiency in large language models during reasoning tasks. Our approach uses Shannon entropy from token-level logprobs as a confidence signal to enable early stopping, achieving 25-50% computational savings while maintaining ta…

• Says Who? Effective Zero-Shot Annotation of Focalization

  Rebecca M. M. Hicke, Yuri Bizzoni, Pascale Feldkamp, Ross Deans Kristensen-McLachlan · 29. Oktober 2025

  Focalization describes the way in which access to narrative information is restricted or controlled based on the knowledge available to knowledge of the narrator. It is encoded via a wide range of lexico-grammatical features and is subject to reader interpretation. Even trained annotators frequently…