Computation and Language
Investigating Disability Representations in Text-to-Image Models
Yang Tian, Yu Fan, Liudmila Zavolokina et al. (+1 more)
Text-to-image generative models have made remarkable progress in producing high-quality visual content from textual descriptions, yet concerns remain about how they represent social groups. While characteristics like gender and race have received increasing
attention, disability representations remain underexplored. This study investigates how people with disabilities are represented in AI-generated images by analyzing outputs from Stable
Diffusion XL and DALL-E 3 using a structured
prompt design. We analyze disability representations by comparing image similarities between generic disability
prompts and
prompts referring to specific disability categories. Moreover, we evaluate how mitigation strategies influence disability portrayals, with a focus on assessing affective framing through sentiment polarity analysis, combining both automatic and human evaluation. Our findings reveal persistent representational imbalances and highlight the need for continuous evaluation and refinement of generative models to foster more diverse and inclusive portrayals of disability.
Read the full abstract ▾
All authors Yang Tian, Yu Fan, Liudmila Zavolokina, Sarah Ebling
Computation and Language · Cross-listed from cs.SD
Audio ControlNet for Fine-Grained Audio Generation and Editing
Haina Zhu, Yao Xiao, Xiquan Li et al. (+5 more)
We study the fine-grained text-to-audio (T2A) generation task. While recent models can synthesize high-quality audio from text descriptions, they often lack precise control over attributes such as loudness, pitch, and sound events. Unlike prior approaches that retrain models for specific control types, we propose to train ControlNet models on top of pre-trained T2A backbones to achieve controllable generation over loudness, pitch, and event roll. We introduce two designs, T2A-ControlNet and T2A-Adapter, and show that the T2A-Adapter model offers a more efficient structure with strong control ability. With only 38M additional parameters, T2A-Adapter achieves state-of-the-art performance on the AudioSet-Strong in both event-level and segment-level F1 scores. We further extend this framework to audio editing, proposing T2A-Editor for removing and inserting audio events at time locations specified by instructions. Models, code,
dataset pipelines, and
benchmarks will be released to support future research on controllable audio generation and editing.
Read the full abstract ▾
All authors Haina Zhu, Yao Xiao, Xiquan Li, Ziyang Ma, Jianwei Yu, Bowen Zhang, Mingqi Yang, Xie Chen
Computation and Language · Cross-listed from cs.SI
Overstating Attitudes, Ignoring Networks: LLM Biases in Simulating Misinformation Susceptibility
Eun Cheol Choi, Lindsay E. Young and Emilio Ferrara
Large
language models (LLMs) are increasingly used as proxies for human judgment in computational social science, yet their ability to reproduce patterns of susceptibility to misinformation remains unclear. We test whether LLM-simulated survey respondents, prompted with participant profiles drawn from social survey data measuring network, demographic, attitudinal and behavioral features, can reproduce human patterns of misinformation belief and sharing. Using three online surveys as baselines, we evaluate whether LLM outputs match observed response distributions and recover feature-outcome associations present in the original survey data. LLM-generated responses capture broad distributional tendencies and show modest correlation with human responses, but consistently overstate the association between belief and sharing. Linear models fit to simulated responses exhibit substantially higher explained variance and place disproportionate weight on attitudinal and behavioral features, while largely ignoring personal network characteristics, relative to models fit to human responses. Analyses of model-generated
reasoning and LLM training data suggest that these distortions reflect systematic biases in how misinformation-related concepts are represented. Our findings suggest that LLM-based survey simulations are better suited for diagnosing systematic divergences from human judgment than for substituting it.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.LG
Delving into Muon and Beyond: Deep Analysis and Extensions
Xianbiao Qi, Marco Chen, Jiaquan Ye et al. (+2 more)
The
Muon optimizer has recently attracted considerable
attention for its strong empirical performance and use of orthogonalized updates on
matrix-shaped parameters, yet its underlying mechanisms and relationship to adaptive optimizers such as Adam remain insufficiently understood. In this work, we aim to address these questions through a unified spectral perspective. Specifically, we view
Muon as the p = 0 endpoint of a family of spectral transformations of the form U \boldsymbol{\Sigma}^{p} V' , and consider additional variants with p = 1/2 , p = 1/4 , and p = 1 . These transformations are applied to both first-moment updates, as in momentum SGD, and to root-mean-square (RMS) normalized gradient updates as in Adam. To enable efficient computation, we develop a coupled Newton iteration that avoids explicit singular value decomposition. Across controlled experiments, we find that RMS-normalized updates yield more stable optimization than first-moment updates. Moreover, while spectral compression provides strong stabilization benefits under first-moment updates, the
Muon update (p = 0) does not consistently outperform Adam. These results suggest that
Muon is best understood as an effective form of spectral normalization, but not a universally superior optimization method. Our source code will be released at https://github.com/Ocram7/BeyondMuon.
Read the full abstract ▾
All authors Xianbiao Qi, Marco Chen, Jiaquan Ye, Yelin He, Rong Xiao
Computation and Language
Approaches to Semantic Textual Similarity in Slovak Language: From Algorithms to Transformers
Lukas Radosky, Miroslav Blstak, Matej Krajcovic et al. (+1 more)
Semantic textual similarity (STS) plays a crucial role in many natural
language processing tasks. While extensively studied in high-resource
languages, STS remains challenging for under-resourced
languages such as Slovak. This paper presents a comparative evaluation of sentence-level STS methods applied to Slovak, including traditional
algorithms, supervised machine learning models, and third-party deep learning tools. We trained several machine learning models using outputs from traditional
algorithms as features, with feature selection and hyperparameter tuning jointly guided by artificial bee colony optimization. Finally, we evaluated several third-party tools, including fine-tuned model by CloudNLP, OpenAI's
embedding models, GPT-4 model, and pretrained SlovakBERT model. Our findings highlight the trade-offs between different approaches.
Read the full abstract ▾
All authors Lukas Radosky, Miroslav Blstak, Matej Krajcovic, Ivan Polasek
Computation and Language
Outcome Accuracy is Not Enough: Aligning the Reasoning Process of Reward Models
Binghai Wang, Yantao Liu, Yuxuan Liu et al. (+13 more)
Generative Reward Models (GenRMs) and LLM-as-a-Judge exhibit deceptive
alignment by producing correct judgments for incorrect reasons, as they are trained and evaluated to prioritize Outcome Accuracy, which undermines their ability to generalize during RLHF. We introduce Rationale Consistency, a fine-grained metric that quantifies the
alignment between the model's
reasoning process and human judgment. Our evaluation of frontier models reveals that rationale consistency effectively discriminates among state-of-the-art models and detects deceptive
alignment, while outcome accuracy falls short in both respects. To mitigate this gap, we introduce a hybrid signal that combines rationale consistency with outcome accuracy for GenRM training. Our training method achieves state-of-the-art performance on RM-Bench (87.1%) and JudgeBench (82%), surpassing outcome-only baselines by an average of 5%. Using RM during RLHF, our method effectively improves performance as demonstrated on Arena Hard v2, notably yielding a 7% improvement in creative writing tasks. Further analysis confirms that our method escapes the deceptive
alignment trap, effectively reversing the decline in rationale consistency observed in outcome-only training.
Read the full abstract ▾
All authors Binghai Wang, Yantao Liu, Yuxuan Liu, Tianyi Tang, Shenzhi Wang, Chang Gao, Chujie Zheng, Yichang Zhang, Le Yu, Shixuan Liu, Tao Gui, Qi Zhang, Xuanjing Huang, Bowen Yu, Fei Huang, Junyang Lin
Computation and Language
Mapping the Web of Science, a large-scale graph and text-based dataset with LLM embeddings
Tim Kunt, Annika Buchholz, Imene Khebouri et al. (+3 more)
Large text data sets, such as publications, websites, and other text-based media, inherit two distinct types of features: (1) the text itself, its information conveyed through semantics, and (2) its relationship to other texts through links, references, or shared attributes. While the latter can be described as a graph structure and can be handled by a range of established
algorithms for classification and prediction, the former has recently gained new potential through the use of LLM
embedding models. Demonstrating these possibilities and their practicability, we investigate the Web of Science
dataset, containing ~56 million scientific publications through the lens of our proposed
embedding method, revealing a self-structured landscape of texts.
Read the full abstract ▾
All authors Tim Kunt, Annika Buchholz, Imene Khebouri, Thorsten Koch, Ida Litzel, Thi Huong Vu
Computation and Language
LEAD: Layer-wise Expert-aligned Decoding for Faithful Radiology Report Generation
Ruixiao Yang, Yuanhe Tian, Di Dong et al. (+3 more)
Radiology Report Generation aims to produce accurate and coherent diagnostics from medical images. Although large
vision-
language models improve report fluency and accuracy, they still suffer from
hallucinations by generating plausible pathological descriptions that are not supported by the input images. Existing methods primarily rely on external knowledge guidance to facilitate the
alignment between generated text and visual information. However, these approaches often ignore the inherent decoding priors and
vision-
language alignment biases in pretrained models and lack robustness due to reliance on constructed guidance. In this paper, we propose Layer-wise Expert-aligned Decoding, a method that directly intervenes in the internal decoding process of large
vision-
language models. A pathology-specific expert module is designed to extract discriminative pathological features, which are then injected into each decoder layer through a gated mechanism. This architecture enables the large
language model to progressively incorporate expert features during generation through a learned layer-wise gating function, thereby mitigating decoding biases and steering generation toward factual consistency. Experiments on multiple public
datasets demonstrate that the proposed method improves clinical accuracy and factual consistency while maintaining competitive report generation quality.
Read the full abstract ▾
All authors Ruixiao Yang, Yuanhe Tian, Di Dong, Xu Yang, Huiqi Li, Yan Song
Computation and Language
Translation Heads: Disentangling meaning from language in LLM-based machine translation
Théo Lasnier, Armel Zebaze, Djamé Seddah et al. (+2 more)
Mechanistic Interpretability (MI) seeks to explain how neural networks implement their capabilities, but the scale of Large
Language Models (LLMs) has limited prior MI work in Machine Translation (MT) to word-level analyses. We study sentence-level MT from a mechanistic perspective by analyzing
attention heads to understand how LLMs internally encode and distribute translation functions. We decompose MT into two subtasks: producing text in the target
language (i.e. target
language identification) and preserving the input sentence's meaning (i.e. sentence equivalence). Across three families of open-source models and 20 translation directions, we find that distinct, sparse sets of
attention heads specialize in each subtask. Based on this insight, we construct subtask-specific steering vectors and show that modifying just 1% of the relevant heads enables instruction-free MT performance comparable to instruction-based prompting, while ablating these heads selectively disrupts their corresponding translation functions.
Read the full abstract ▾
All authors Théo Lasnier, Armel Zebaze, Djamé Seddah, Rachel Bawden, Benoît Sagot
Computation and Language
Focus-LIME: Surgical Interpretation of Long-Context Large Language Models via Proxy-Based Neighborhood Selection
Junhao Liu, Haonan Yu, Zhenyu Yan et al. (+1 more)
As Large
Language Models (LLMs) scale to handle massive context windows, achieving
surgical feature-level interpretation is essential for high-stakes tasks like legal
auditing and code debugging. However, existing local model-agnostic explanation methods face a critical dilemma in these scenarios: feature-based methods suffer from attribution dilution due to high feature dimensionality, thus failing to provide faithful explanations. In this paper, we propose Focus-LIME, a coarse-to-fine framework designed to restore the tractability of
surgical interpretation. Focus-LIME utilizes a proxy model to curate the perturbation neighborhood, allowing the target model to perform fine-grained attribution exclusively within the optimized context. Empirical evaluations on long-context
benchmarks demonstrate that our method makes
surgical explanations practicable and provides faithful explanations to users.
Read the full abstract ▾
All authors Junhao Liu, Haonan Yu, Zhenyu Yan, Xin Zhang
Computation and Language
RexBERT: Context Specialized Bidirectional Encoders for E-commerce
Rahul Bajaj and Anuj Garg
Encoder-only
transformers remain indispensable in
retrieval, classification, and ranking systems where latency, stability, and cost are paramount. Most general purpose encoders, however, are trained on generic corpora with limited coverage of specialized domains. We introduce RexBERT, a family of BERT-style encoders designed specifically for e-commerce semantics. We make three contributions. First, we release Ecom-niverse, a 350 billion
token corpus curated from diverse retail and shopping sources. We describe a modular pipeline that isolates and extracts e-commerce content from FineFineWeb and other open web resources, and characterize the resulting domain distribution. Second, we present a reproducible pretraining recipe building on ModernBERT's architectural advances. The recipe consists of three phases: general pre-training, context extension, and annealed domain specialization. Third, we train RexBERT models ranging from 17M to 400M parameters and evaluate them on
token classification, semantic similarity, and general natural
language understanding tasks using e-commerce
datasets. Despite having 2-3x fewer parameters, RexBERT outperforms larger general-purpose encoders and matches or surpasses modern long-context models on domain-specific
benchmarks. Our results demonstrate that high quality in-domain data combined with a principled training approach provides a stronger foundation for e-commerce applications than indiscriminate scaling alone.
Read the full abstract ▾
Computation and Language
Beyond Holistic Scores: Automatic Trait-Based Quality Scoring of Argumentative Essays
Lucile Favero, Juan Antonio Pérez-Ortiz, Tanja Käser et al. (+1 more)
Automated Essay Scoring systems have traditionally focused on holistic scores, limiting their pedagogical usefulness, especially in the case of complex essay genres such as argumentative writing. In educational contexts, teachers and learners require interpretable, trait-level feedback that aligns with instructional goals and established rubrics. In this paper, we study trait-based Automatic Argumentative Essay Scoring using two complementary modeling paradigms designed for realistic educational deployment: (1) structured in-context learning with small open-source LLMs, and (2) a supervised, encoder-based BigBird model with a CORAL-style ordinal regression formulation, optimized for long-sequence understanding. We conduct a systematic evaluation on the ASAP++
dataset, which includes essay scores across five quality traits, offering strong coverage of core argumentation dimensions. LLMs are prompted with designed, rubric-aligned in-context examples, along with feedback and confidence requests, while we explicitly model ordinality in scores with the BigBird model via the rank-consistent CORAL framework. Our results show that explicitly modeling score ordinality substantially improves agreement with human raters across all traits, outperforming LLMs and nominal classification and regression-based baselines. This finding reinforces the importance of aligning model objectives with rubric semantics for educational assessment. At the same time, small open-source LLMs achieve a competitive performance without task-specific fine-tuning, particularly for
reasoning-oriented traits, while enabling transparent,
privacy-preserving, and locally deployable assessment scenarios. Our findings provide methodological, modeling, and practical insights for the design of AI-based educational systems that aim to deliver interpretable, rubric-aligned feedback for argumentative writing.
Read the full abstract ▾
All authors Lucile Favero, Juan Antonio Pérez-Ortiz, Tanja Käser, Nuria Oliver
Computation and Language
VILLAIN at AVerImaTeC: Verifying Image-Text Claims via Multi-Agent Collaboration
Jaeyoon Jung, Yejun Yoon and Kunwoo Park
This paper describes VILLAIN, a
multimodal fact-checking system that verifies image-text claims through
prompt-based multi-
agent collaboration. For the AVerImaTeC shared task, VILLAIN employs
vision-
language model
agents across multiple stages of fact-checking. Textual and visual evidence is retrieved from the knowledge store enriched through additional web collection. To identify key information and address inconsistencies among evidence items, modality-specific and cross-modal
agents generate analysis reports. In the subsequent stage, question-answer pairs are produced based on these reports. Finally, the Verdict Prediction
agent produces the verification outcome based on the image-text claim and the generated question-answer pairs. Our system ranked first on the leaderboard across all evaluation metrics. The source code is publicly available at https://github.com/ssu-humane/VILLAIN.
Read the full abstract ▾
Computation and Language
Trust The Typical
Debargha Ganguly, Sreehari Sankar, Biyao Zhang et al. (+8 more)
Current approaches to LLM safety fundamentally rely on a brittle cat-and-mouse game of identifying and blocking known threats via guardrails. We argue for a fresh approach: robust safety comes not from enumerating what is harmful, but from deeply understanding what is safe. We introduce Trust The Typical (T3), a framework that operationalizes this principle by treating safety as an out-of-distribution (OOD) detection problem. T3 learns the distribution of acceptable
prompts in a semantic space and flags any significant deviation as a potential threat. Unlike prior methods, it requires no training on harmful examples, yet achieves state-of-the-art performance across 18
benchmarks spanning toxicity, hate
speech, jailbreaking,
multilingual harms, and over-refusal, reducing false positive rates by up to 40x relative to specialized safety models. A single model trained only on safe English text transfers effectively to diverse domains and over 14
languages without retraining. Finally, we demonstrate production readiness by integrating a GPU-optimized version into vLLM, enabling continuous guardrailing during
token generation with less than 6% overhead even under dense evaluation intervals on large-scale workloads.
Read the full abstract ▾
All authors Debargha Ganguly, Sreehari Sankar, Biyao Zhang, Vikash Singh, Kanan Gupta, Harshini Kavuru, Alan Luo, Weicong Chen, Warren Morningstar, Raghu Machiraju, Vipin Chaudhary
Computation and Language · Cross-listed from cs.IR
AIANO: Enhancing Information Retrieval with AI-Augmented Annotation
Sameh Khattab, Marie Bauer, Lukas Heine et al. (+3 more)
The rise of Large
Language Models (LLMs) and
Retrieval-Augmented Generation (RAG) has rapidly increased the need for high-quality, curated information
retrieval datasets. These
datasets, however, are currently created with off-the-shelf annotation tools that make the annotation process complex and inefficient. To streamline this process, we developed a specialized annotation tool - AIANO. By adopting an AI-augmented annotation workflow that tightly integrates human expertise with LLM assistance, AIANO enables annotators to leverage AI suggestions while retaining full control over annotation decisions. In a within-subject user study ($n = 15$), participants created question-answering
datasets using both a baseline tool and AIANO. AIANO nearly doubled annotation speed compared to the baseline while being easier to use and improving
retrieval accuracy. These results demonstrate that AIANO's AI-augmented approach accelerates and enhances
dataset creation for information
retrieval tasks, advancing annotation capabilities in
retrieval-intensive domains.
Read the full abstract ▾
All authors Sameh Khattab, Marie Bauer, Lukas Heine, Till Rostalski, Jens Kleesiek, Julian Friedrich
Computation and Language
Edward Phillips, Sean Wu, Fredrik K. Gustafsson et al. (+2 more)
Large
language models present challenges for principled uncertainty quantification, in part due to their complexity and the diversity of their outputs. Semantic dispersion, or the variance in the meaning of sampled answers, has been proposed as a useful proxy for model uncertainty, but the associated computational cost prohibits its use in latency-critical applications. We show that sampled semantic distributions can be distilled into lightweight student models which estimate a
prompt-conditioned density before the
language model generates an answer
token. The student model predicts a semantic distribution over possible answers; the
entropy of this distribution provides a
prompt-level uncertainty signal, and the probability density allows answer-level reliability evaluation. Across experiments on TriviaQA and MMLU, we find our student models perform competitively relative to the teacher's sampled semantic dispersion on a
hallucination prediction task, whilst offering additional uncertainty primitives for out-of-domain detection and multiple-choice answer selection. We term this technique Semantic Self-
Distillation (SSD), which can serve as a general framework for distilling predictive uncertainty in complex output spaces beyond
language.
Read the full abstract ▾
All authors Edward Phillips, Sean Wu, Fredrik K. Gustafsson, Boyan Gao, David A. Clifton
Computation and Language
Can LLMs capture stable human-generated sentence entropy measures?
Estrella Pivel-Villanueva, Elisabeth Frederike Sterner and Franziska Knolle
Predicting upcoming words is a core mechanism of
language comprehension and may be quantified using Shannon
entropy. There is currently no empirical consensus on how many human responses are required to obtain stable and unbiased
entropy estimates at the word level. Moreover, large
language models (LLMs) are increasingly used as substitutes for human norming data, yet their ability to reproduce stable human
entropy remains unclear. Here, we address both issues using two large publicly available cloze
datasets in German 1 and English 2. We implemented a bootstrap-based convergence analysis that tracks how
entropy estimates stabilize as a function of sample size. Across both
languages, more than 97% of sentences reached stable
entropy estimates within the available sample sizes. 90% of sentences converged after 111 responses in German and 81 responses in English, while low-
entropy sentences (<1) required as few as 20 responses and high-
entropy sentences (>2.5) substantially more. These findings provide the first direct empirical validation for common norming practices and demonstrate that convergence critically depends on sentence predictability. We then compared stable human
entropy values with
entropy estimates derived from several LLMs, including GPT-4o, using both logit-based probability extraction and sampling-based frequency estimation, GPT2-xl/german-GPT-2, RoBERTa Base/GottBERT, and LLaMA 2 7B Chat. GPT-4o showed the highest correspondence with human data, although
alignment depended strongly on the extraction method and
prompt design. Logit-based estimates minimized absolute error, whereas sampling-based estimates were better in capturing the dispersion of human variability. Together, our results establish practical guidelines for human norming and show that while LLMs can approximate human
entropy, they are not interchangeable with stable human-derived distributions.
Read the full abstract ▾
Computation and Language
Textual Planning with Explicit Latent Transitions
Eliezer Shlomi, Ido Levy, Eilam Shapira et al. (+6 more)
Planning requires a transition model that predicts how each action changes the current state. When a large
language model (LLM) plays this role, every next state is generated
token by
token, which makes searching over many possible futures slow and expensive. Existing alternatives either still query an LLM at every step or require a symbolic model of the domain. We propose EmbedPlan, a transition model built on frozen text
embeddings: it embeds natural
language descriptions of the state and the action with a frozen LLM, predicts the
embedding of the next state with a lightweight learned network, and returns the closest real state. Because this network can be trained on top of any encoder, EmbedPlan also provides a controlled way to compare text representations for learning transitions. We evaluate it on 9 classical planning domains, under six settings that hold out progressively more of the data, from transitions to entire domains, and against baselines ranging from predicting no change to learning symbolic action rules. On planning problems seen during training, EmbedPlan almost always ranks the true next state among its top five guesses, still does so for most queries even when every observed state is a candidate, and retains 92-99% of its single-step accuracy when predicting several steps ahead from its own outputs. Given the same candidate states as GPT-5.4, it picks the true next state more often while taking about 0.17 ms per transition with cached
embeddings. Accuracy is lower on unseen problems and near chance on unseen domains, and the controlled comparison traces this limit to the state representation rather than to the learned transition.
Read the full abstract ▾
All authors Eliezer Shlomi, Ido Levy, Eilam Shapira, Michael Katz, Guy Uziel, Segev Shlomov, Nir Mashkif, Roi Reichart, Sarah Keren
Computation and Language
Rethinking Weight Tying: Pseudo-Inverse Tying for LM Stable Training and Updates
Jian Gu, Aldeida Aleti, Chunyang Chen et al. (+1 more)
Weight tying is widely used in compact
language models to reduce parameters by sharing the
token table between the input
embedding and the output projection. However, parameter sharing alone does not guarantee a stable
token interface: during training, the correspondence between encoding
tokens into hidden states and decoding hidden states into logits can drift, worsening optimization sensitivity and weakening explainability probes that rely on a meaningful vocabulary-space decoder. We propose Pseudo-Inverse Tying (PIT), which synchronizes
embedding and unembedding as coupled projections of a shared latent
token memory, guaranteeing a pseudo-inverse-consistent interface throughout training. PIT maintains an orthonormal shared
memory, obtained by polar initialization from a source checkpoint for continued pretraining or by random orthonormal initialization for from-scratch pretraining, and introduces a learned symmetric positive definite hidden-space transform parameterized via a Cholesky factor. The output head applies this transform to hidden states before the vocabulary projection, while the
embedding applies the inverse transform to
token vectors using stable triangular solves, avoiding explicit pseudo-inverse recomputation and vocabulary-sized auxiliary parameters. Beyond improving training stability, PIT provides a cleaner substrate for logit-lens-style and vocabulary-space explainability probes by keeping the input and output
token geometries synchronized. We evaluate PIT on on-device models spanning 256M-1.3B parameters. The results show that PIT improves continued-pretraining stability, enforces near-exact
token-interface consistency across settings, and yields more predictable lightweight adaptation after continued pretraining, while from-scratch pretraining reveals a trade-off between strict interface consistency and unconstrained optimization.
Read the full abstract ▾
All authors Jian Gu, Aldeida Aleti, Chunyang Chen, Hongyu Zhang
Computation and Language · Cross-listed from cs.SI
Unmasking Superspreaders: Data-Driven Approaches for Identifying and Comparing Key Influencers of Conspiracy Theories on X.com
Florian Kramer, Henrich R. Greve, Moritz von Zahn et al. (+1 more)
Conspiracy theories can threaten society by spreading misinformation, deepening polarization, and eroding trust in democratic institutions. Social media often fuels the spread of conspiracies, primarily driven by two key actors: Superspreaders – influential individuals disseminating conspiracy content at disproportionately high rates, and Bots – automated accounts designed to amplify conspiracies strategically. To counter the spread of conspiracy theories, it is critical to both identify these actors and to better understand their behavior. However, a systematic analysis of these actors as well as real-world-applicable identification methods are still lacking. In this study, we leverage over seven million tweets from the COVID-19 pandemic to analyze key differences between Human Superspreaders and Bots across dimensions such as linguistic complexity, toxicity, and hashtag usage. Our analysis reveals distinct communication strategies: Superspreaders tend to use more complex
language and substantive content while relying less on structural elements like hashtags and emojis, likely to enhance credibility and authority. By contrast, Bots favor simpler
language and strategic cross-usage of hashtags, likely to increase accessibility, facilitate infiltration into trending discussions, and amplify reach. To counter both Human Superspreaders and Bots, we propose and evaluate 27 novel metrics for quantifying the severity of conspiracy theory spread. Our findings highlight the effectiveness of an adapted H-Index for computationally feasible identification of Human Superspreaders. By identifying behavioral patterns unique to Human Superspreaders and Bots as well as providing suitable identification methods, this study provides a foundation for mitigation strategies, including platform moderation policies, temporary and permanent account suspensions, and public awareness campaigns.
Read the full abstract ▾
All authors Florian Kramer, Henrich R. Greve, Moritz von Zahn, Hayagreeva Rao
Computation and Language
LycheeDecode: Accelerating Long-Context LLM Inference via Hybrid-Head Sparse Decoding
Gang Lin, Dongfang Li, Zhuoen Chen et al. (+4 more)
The proliferation of long-context large
language models (LLMs) exposes a key bottleneck: the rapidly expanding key-value cache during decoding, which imposes heavy
memory and latency costs. While recent approaches attempt to alleviate this by sharing a single set of crucial
tokens across layers, such coarse-grained sharing undermines model performance by neglecting the functional diversity of
attention heads. To address this, we propose LycheeDecode, an efficient decoding method centered on a fine-grained hybrid-head
attention mechanism that employs a hardware-efficient top-k selection strategy. Specifically, the novel HardKuma-based mechanism partitions
attention heads into a small subset of
retrieval heads that dynamically identify crucial
tokens and a majority of sparse heads that reuse them for efficient computation. Through extensive experiments on leading models like Llama3 and Qwen3 across diverse
benchmarks for long-context understanding (e.g., LongBench, RULER) and complex
reasoning (e.g., AIME24, OlympiadBench), we demonstrate that LycheeDecode achieves generative quality comparable to, and at times surpassing even the full-
attention baseline. Crucially, this is accomplished with up to a 2.7x speedup at a 128K context length. By preserving the functional diversity of
attention heads, our fine-grained strategy overcomes the performance bottlenecks of existing methods, providing a powerful and validated pathway to both efficient and high-quality long-context LLM inference.
Read the full abstract ▾
All authors Gang Lin, Dongfang Li, Zhuoen Chen, Yukun Shi, Xuhui Chen, Baotian Hu, Min Zhang
Computation and Language · Cross-listed from cs.HC
PersoPilot: An Adaptive AI-Copilot for Transparent Contextualized Persona Classification and Personalized Response Generation
Saleh Afzoon, Amin Beheshti and Usman Naseem
Understanding and classifying user personas is critical for delivering effective personalization. While persona information offers valuable insights, its full potential is realized only when contextualized, linking user characteristics with situational context to enable more precise and meaningful service provision. Existing systems often treat persona and context as separate inputs, limiting their ability to generate nuanced, adaptive interactions. To address this gap, we present PersoPilot, an agentic AI-Copilot that integrates persona understanding with contextual analysis to support both end users and analysts. End users interact through a transparent, explainable chat interface, where they can express preferences in natural
language, request recommendations, and receive information tailored to their immediate task. On the analyst side, PersoPilot delivers a transparent,
reasoning-powered labeling assistant, integrated with an active learning-driven classification process that adapts over time with new labeled data. This feedback loop enables targeted service recommendations and adaptive personalization, bridging the gap between raw persona data and actionable, context-aware insights. As an adaptable framework, PersoPilot is applicable to a broad range of service personalization scenarios.
Read the full abstract ▾
Computation and Language
$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal
Aditya Kasliwal, Pratinav Seth and Vinay Kumar Sankarapu
Modern deployments require LLMs to enforce safety policies at scale, yet many controls rely on inference-time interventions that add recurring compute cost and serving complexity. Activation steering is widely used, but it requires runtime hooks and scales cost with the number of generations; conditional variants improve selectivity by gating when steering is applied but still retain an inference-time control path. We ask whether selective refusal can be moved entirely offline: can a mechanistic understanding of category-specific refusal be distilled into a
circuit-restricted weight update that deploys as a standard checkpoint? We propose C-{\Delta}{\theta}
Circuit Restricted Weight Arithmetic}, which (i) localizes refusal-causal computation as a sparse
circuit using EAP-IG and (ii) computes a constrained weight update {\Delta}{\theta}C supported only on that
circuit (typically <5% of parameters). Applying {\Delta}{\theta}C yields a drop-in edited checkpoint with no inference-time hooks, shifting cost from per request intervention to a one-time offline update. We evaluate category-targeted selectivity and capability retention on refusal and utility
benchmarks.
Read the full abstract ▾
Computation and Language
ReFRAME or Remain: Unsupervised Lexical Semantic Change Detection with Frame Semantics
Bach Phan-Tat, Kris Heylen, Dirk Geeraerts et al. (+2 more)
The majority of contemporary computational methods for lexical semantic change (LSC) detection are based on neural
embedding distributional representations. Although these models perform well on LSC
benchmarks, their results are often difficult to interpret. We explore an alternative approach that relies solely on frame semantics. We show that this method is effective for detecting semantic change and can even outperform many distributional semantic models. Finally, we present a detailed quantitative and qualitative analysis of its predictions, demonstrating that they are both plausible and highly interpretable
Read the full abstract ▾
All authors Bach Phan-Tat, Kris Heylen, Dirk Geeraerts, Stefano De Pascale, Dirk Speelman
Computation and Language
Model-Dowser: Data-Free Importance Probing to Mitigate Catastrophic Forgetting in Multimodal Large Language Models
Hyeontaek Hwang, Nguyen Dinh Son and Daeyoung Kim
Fine-tuning
Multimodal Large
Language Models (MLLMs) on task-specific data is an effective way to improve performance on downstream applications. However, such adaptation often leads to a degradation in generalization on pretrained tasks, a phenomenon known as Catastrophic Forgetting. Existing methods that aim to mitigate this issue either become ineffective when fine-tuning deeper layers of the
language decoder or scale poorly with increasing model size. To address these limitations, we propose Model-Dowser, a novel sparse fine-tuning approach for MLLMs. Model-Dowser measures a principled importance score for each model parameter with respect to pretrained generalization (prior to downstream adaptation) by jointly considering weight magnitudes, input activations, and output sensitivities. During fine-tuning, Model-Dowser selectively preserves high-importance parameters and updates the remaining. Comprehensive experiments on two representative MLLMs, LLaVA and NVILA, demonstrate that Model-Dowser effectively mitigates catastrophic forgetting and consistently outperforms prior methods, while remaining resource-efficient and scalable to multi-billion-parameter models.
Read the full abstract ▾
Computation and Language
PersoDPO: Scalable Preference Optimization for Instruction-Adherent, Persona-Grounded Dialogue via Multi-LLM Evaluation
Saleh Afzoon, MohammadHossein Ahmadi, Usman Naseem et al. (+1 more)
Personalization and contextual coherence are two essential components in building effective persona-grounded dialogue systems. These aspects play a crucial role in enhancing user engagement and ensuring responses are more relevant and consistent with user identity. However, recent studies indicate that open-source large
language models (LLMs) continue to struggle to generate responses that are both contextually grounded and aligned with persona cues, despite exhibiting strong general conversational abilities like fluency and naturalness. We present PersoDPO, a scalable preference optimisation framework that uses supervision signals from automatic evaluations of responses generated by both closed-source and open-source LLMs to fine-tune dialogue models. The framework integrates evaluation metrics targeting coherence and personalization, along with a length-format compliance feature to promote instruction adherence. These signals are combined to automatically construct high-quality preference pairs without manual annotation, enabling a scalable and reproducible training pipeline. Experiments on the FoCus
dataset show that an open-source
language model fine-tuned with the PersoDPO framework consistently outperforms strong open-source baselines and a standard Direct Preference Optimization (DPO) variant across multiple evaluation dimensions.
Read the full abstract ▾
All authors Saleh Afzoon, MohammadHossein Ahmadi, Usman Naseem, Amin Beheshti
Computation and Language
LLM surprisal is necessary but not sufficient to capture English garden-path effects: Evidence from joint latent modeling of reading paradigms
Dario Paape, Tal Linzen and Shravan Vasishth
Temporarily ambiguous garden-path sentences ("While the team trained the striker wondered... ") are known to cause processing difficulty, which can manifest itself in a variety of reading behaviors (in-situ slowdowns, rereading), as well as in miscomprehension or outright rejection of the sentence as ungrammatical. Which types of reading behavior are observed critically depends on the experimental method used to collect the data, which makes comparing results between reading paradigms difficult. To address this problem, we present a latent-process multinomial processing tree (MPT) model of human reading and comprehension/judgment behavior in garden-path sentences that we fit to combined data from four different reading paradigms (eye tracking, uni- and bidirectional self-paced reading, Maze). The model distinguishes between the probability of adopting an incorrect initial analysis, the cost of encountering an incompatible continuation, and the cost of syntactic reanalysis. By taking into account trials with inattentive reading, more realistic estimates of the cost parameters are obtained. Cross-validation reveals that the MPT model has a better predictive fit to human reading patterns and end-of-trial task data than a model based solely on LLM-derived surprisal values. We also test several models that assume an influence of surprisal within the MPT architecture, and find that adding surprisal as an additional predictor or reading time and/or garden-path cost further improves predictive fit.
Read the full abstract ▾
Computation and Language
Beyond Unimodal Shortcuts: MLLMs as Cross-Modal Reasoners for Grounded Named Entity Recognition
Jinlong Ma, Yu Zhang, Xuefeng Bai et al. (+5 more)
Grounded
Multimodal Named Entity Recognition (GMNER) aims to extract text-based entities, assign them semantic categories, and ground them to corresponding visual regions. In this work, we explore the potential of
Multimodal Large
Language Models (MLLMs) to perform GMNER in an end-to-end manner, moving beyond their typical role as auxiliary tools within cascaded pipelines. Crucially, our investigation reveals a fundamental challenge: MLLMs exhibit $\textbf{modality bias}$, including visual bias and textual bias, which stems from their tendency to take unimodal shortcuts rather than rigorous cross-modal verification. To address this, we propose Modality-aware Consistency
Reasoning ($\textbf{MCR}$), which enforces structured cross-modal
reasoning through Multi-style
Reasoning Schema Injection (MRSI) and Constraint-guided Verifiable Optimization (CVO). MRSI transforms abstract constraints into executable
reasoning chains, while CVO empowers the model to dynamically align its
reasoning trajectories with Group Relative Policy Optimization (GRPO). Experiments on GMNER and visual grounding tasks demonstrate that MCR effectively mitigates modality bias and achieves superior performance compared to existing baselines.
Read the full abstract ▾
All authors Jinlong Ma, Yu Zhang, Xuefeng Bai, Kehai Chen, Yuwei Wang, Zeming Liu, Jun Yu, Min Zhang
Computation and Language
Is Micro Domain-Adaptive Pre-Training Effective for Real-World Operations? Multi-Step Evaluation Reveals Potential and Bottlenecks
Masaya Tsunokake, Yuta Koreeda, Terufumi Morishita et al. (+3 more)
When applying LLMs to real-world enterprise operations, LLMs need to handle proprietary knowledge in small domains of specific operations ($\textbf{micro domains}$). A previous study shows micro domain-adaptive pre-training ($\textbf{mDAPT}$) with fewer documents is effective, similarly to DAPT in larger domains. However, it evaluates mDAPT only on multiple-choice questions; thus, its effectiveness for generative tasks in real-world operations remains unknown. We aim to reveal the potential and bottlenecks of mDAPT for generative tasks. To this end, we disentangle the answering process into three subtasks and evaluate the performance of each subtask: (1) $\textbf{eliciting}$ facts relevant to questions from an LLM's own knowledge, (2) $\textbf{reasoning}$ over the facts to obtain conclusions, and (3) $\textbf{composing}$ long-form answers based on the conclusions. We verified mDAPT on proprietary IT product knowledge for real-world questions in IT technical support operations. As a result, mDAPT resolved the elicitation task that the base model struggled with but did not resolve other subtasks. This clarifies mDAPT's effectiveness in the knowledge aspect and its bottlenecks in other aspects. Further analysis empirically shows that resolving the elicitation and
reasoning tasks ensures sufficient performance (over 90%), emphasizing the need to enhance
reasoning capability.
Read the full abstract ▾
All authors Masaya Tsunokake, Yuta Koreeda, Terufumi Morishita, Koichi Nagatsuka, Hikaru Tomonari, Yasuhiro Sogawa
Computation and Language · Cross-listed from cs.CY
Growth First, Care Second? Tracing the Landscape of LLM Value Preferences in Everyday Dilemmas
Zhiyi Chen, Eun Cheol Choi, Yingjia Luo et al. (+4 more)
People increasingly seek advice online from both human peers and large
language model (LLM)-based chatbots. Such advice rarely involves identifying a single correct answer; instead, it typically requires navigating trade-offs among competing values. We aim to characterize how LLMs navigate value trade-offs across different advice-seeking contexts. First, we examine the value trade-off structure underlying advice seeking using a curated
dataset from four advice-oriented subreddits. Using a bottom-up approach, we inductively construct a hierarchical value framework by aggregating fine-grained values extracted from individual advice options into higher-level value categories. We construct value co-occurrence networks to characterize how values co-occur within dilemmas and find substantial heterogeneity in value trade-off structures across advice-seeking contexts: a women-focused subreddit exhibits the highest network density, indicating more complex value conflicts; women's, men's, and friendship-related subreddits exhibit highly correlated value-conflict patterns centered on security-related tensions (security vs. respect/connection/commitment); by contrast, career advice forms a distinct structure where security frequently clashes with self-actualization and growth. We then evaluate LLM value preferences against these dilemmas and find that, across models and contexts, LLMs consistently prioritize values related to Exploration & Growth over Benevolence & Connection. This systemically skewed value orientation highlights a potential risk of value homogenization in AI-mediated advice, raising concerns about how such systems may shape decision-making and normative outcomes at scale.
Read the full abstract ▾
All authors Zhiyi Chen, Eun Cheol Choi, Yingjia Luo, Xinyi Wang, Yulei Xiao, Aizi Yang, Luca Luceri