Here is today's selection of cs.CL papers. The common theme among these works is the enhancement of language models through innovative approaches, such as improving counterfactual reasoning in vision-language models, refining tokenization techniques, and exploring the geometry of decision-making. Additionally, advancements in understanding emotional nuances in multiword expressions and developing retrieval-based methods for mispronunciation detection highlight the diverse applications of language processing technologies.

CounterVQA: Evaluating and Improving Counterfactual Reasoning in Vision-Language Models for Video Understanding

Vision Language Models (VLMs) have recently shown significant advancements in video understanding, especially in feature alignment, event reasoning, and instruction-following tasks. However, their capability for counterfactual reasoning, inferring alternative outcomes under hypothetical conditions, remains underexplored. This capability is essential for robust video understanding, as it requires identifying underlying causal structures and reasoning about unobserved possibilities, rather than merely recognizing observed patterns. To systematically evaluate this capability, we introduce CounterVQA, a video-based benchmark featuring three progressive difficulty levels that assess different aspects of counterfactual reasoning. Through comprehensive evaluation of both state-of-the-art open-source and closed-source models, we uncover a substantial performance gap: while these models achieve reasonable accuracy on simple counterfactual questions, performance degrades significantly on complex multi-hop causal chains. To address these limitations, we develop a post-training method, CFGPT, that enhances a model's visual counterfactual reasoning ability by distilling its counterfactual reasoning capability from the language modality, yielding consistent improvements across all CounterVQA difficulty levels. Dataset and code will be further released.

Breaking Bad: Norms for Valence, Arousal, and Dominance for over 10k English Multiword Expressions

Factor analysis studies have shown that the primary dimensions of word meaning are Valence (V), Arousal (A), and Dominance (D). Existing lexicons such as the NRC VAD Lexicon, published in 2018, include VAD association ratings for words. Here, we present a complement to it, which has human ratings of valence, arousal, and dominance for 10k English Multiword Expressions (MWEs) and their constituent words. We also increase the coverage of unigrams, especially words that have become more common since 2018. In all, the new NRC VAD Lexicon v2 now has entries for 10k MWEs and 25k words, in addition to the entries in v1. We show that the associations are highly reliable. We use the lexicon to examine emotional characteristics of MWEs, including: 1. The degree to which MWEs (idioms, noun compounds, and verb particle constructions) exhibit strong emotionality; 2. The degree of emotional compositionality in MWEs. The lexicon enables a wide variety of research in NLP, Psychology, Public Health, Digital Humanities, and Social Sciences. The NRC VAD Lexicon v2 is freely available through the project webpage: http://saifmohammad.com/WebPages/nrc-vad.html

Mispronunciation Detection and Diagnosis Without Model Training: A Retrieval-Based Approach

Mispronunciation Detection and Diagnosis (MDD) is crucial for language learning and speech therapy. Unlike conventional methods that require scoring models or training phoneme-level models, we propose a novel training-free framework that leverages retrieval techniques with a pretrained Automatic Speech Recognition model. Our method avoids phoneme-specific modeling or additional task-specific training, while still achieving accurate detection and diagnosis of pronunciation errors. Experiments on the L2-ARCTIC dataset show that our method achieves a superior F1 score of 69.60% while avoiding the complexity of model training.

Length-MAX Tokenizer for Language Models

We introduce a new tokenizer for language models that minimizes the average tokens per character, thereby reducing the number of tokens needed to represent text during training and to generate text during inference. Our method, which we refer to as the Length-MAX tokenizer, obtains its vocabulary by casting a length-weighted objective maximization as a graph partitioning problem and developing a greedy approximation algorithm. On FineWeb and diverse domains, it yields 14--18\% fewer tokens than Byte Pair Encoding (BPE) across vocabulary sizes from 10K to 50K, and the reduction is 13.0\% when the size is 64K. Training GPT-2 models at 124M, 355M, and 1.3B parameters from scratch with five runs each shows 18.5\%, 17.2\%, and 18.5\% fewer steps, respectively, to reach a fixed validation loss, and 13.7\%, 12.7\%, and 13.7\% lower inference latency, together with a 16\% throughput gain at 124M, while consistently improving on downstream tasks including reducing LAMBADA perplexity by 11.7\% and enhancing HellaSwag accuracy by 4.3\%. Moreover, the Length-MAX tokenizer achieves 99.62\% vocabulary coverage and the out-of-vocabulary rate remains low at 0.12\% on test sets. These results demonstrate that optimizing for average token length, rather than frequency alone, offers an effective approach to more efficient language modeling without sacrificing -- and often improving -- downstream performance. The tokenizer is compatible with production systems and reduces embedding and KV-cache memory by 18\% at inference.

Geometry of Decision Making in Language Models

Large Language Models (LLMs) show strong generalization across diverse tasks, yet the internal decision-making processes behind their predictions remain opaque. In this work, we study the geometry of hidden representations in LLMs through the lens of \textit{intrinsic dimension} (ID), focusing specifically on decision-making dynamics in a multiple-choice question answering (MCQA) setting. We perform a large-scale study, with 28 open-weight transformer models and estimate ID across layers using multiple estimators, while also quantifying per-layer performance on MCQA tasks. Our findings reveal a consistent ID pattern across models: early layers operate on low-dimensional manifolds, middle layers expand this space, and later layers compress it again, converging to decision-relevant representations. Together, these results suggest LLMs implicitly learn to project linguistic inputs onto structured, low-dimensional manifolds aligned with task-specific decisions, providing new geometric insights into how generalization and reasoning emerge in language models.