Peer Revenue Fully funded

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science Computation and Language

cs.CL Watch Computation and Language

Saturday 22 November

Computation and Language

SPINE: Token-Selective Test-Time Reinforcement Learning with Entropy-Band Regularization

Jianghao Wu, Yasmeen George, Jin Ye et al. (+3 more)

Large language models (LLMs) and multimodal LLMs (MLL-Ms) excel at chain-of-thought reasoning but face distribution shift at test-time and a lack of verifiable supervision. Recent test-time reinforcement learning (TTRL) methods derive label-free pseudo-rewards from self-consistency voting over sampled trajectories, yet they often collapse: the majority-vote reward prevails, responses shorten, and Pass@1 declines. We trace this to uniform sequence updates in which most tokens are low-entropy followers, while a small high-entropy subset determines the reasoning branches. Thus we propose \method, a token-selective test-time reinforcement learning framework that (i) performs distribution-aware forking-token selection to update only decision-critical branch points, and (ii) applies a robust entropy-band regularizer at those tokens to prevent premature collapse and suppress noisy drift. \method plugs into GRPO-style objectives (optionally with a KL anchor) and requires neither labels nor reward models. Across eight benchmarks spanning multimodal VQA, text-only reasoning, \method consistently improves Pass@1 over TTRL while avoiding response-length collapse and yielding more stable training dynamics on both LLM and MLLM backbones. These results indicate that aligning updates with chain-of-thought branch points is a simple and label-free mechanism for stable and effective test-time adaptation in reasoning models. Code will be released.
Read the full abstract

All authors Jianghao Wu, Yasmeen George, Jin Ye, Yicheng Wu, Daniel F. Schmidt, Jianfei Cai

Computation and Language

Towards Efficient LLM-aware Heterogeneous Graph Learning

Wenda Li, Tongya Zheng, Shunyu Liu et al. (+7 more)

Heterogeneous graphs are widely present in real-world complex networks, where the diversity of node and relation types leads to complex and rich semantics. Efforts for modeling complex relation semantics in heterogeneous graphs are restricted by the limitations of predefined semantic dependencies and the scarcity of supervised signals. The advanced pre-training and fine-tuning paradigm leverages graph structure to provide rich self-supervised signals, but introduces semantic gaps between tasks. Large Language Models (LLMs) offer significant potential to address the semantic issues of relations and tasks in heterogeneous graphs through their strong reasoning capabilities in textual modality, but their incorporation into heterogeneous graphs is largely limited by computational complexity. Therefore, in this paper, we propose an Efficient LLM-Aware (ELLA) framework for heterogeneous graphs, addressing the above issues. To capture complex relation semantics, we propose an LLM-aware Relation Tokenizer that leverages LLM to encode multi-hop, multi-type relations. To reduce computational complexity, we further employ a Hop-level Relation Graph Transformer, which help reduces the complexity of LLM-aware relation reasoning from exponential to linear. To bridge semantic gaps between pre-training and fine-tuning tasks, we introduce the fine-grained task-aware textual Chain-of-Thought (CoT) prompts. Extensive experiments on four heterogeneous graphs show that our proposed ELLA outperforms state-of-the-art methods in the performance and efficiency. In particular, ELLA scales up to 13b-parameter LLMs and achieves up to a 4x speedup compared with existing LLM-based methods. Our code is publicly available at https://github.com/l-wd/ELLA.
Read the full abstract

All authors Wenda Li, Tongya Zheng, Shunyu Liu, Yu Wang, Kaixuan Chen, Hanyang Yuan, Bingde Hu, Zujie Ren, Mingli Song, Gang Chen

Computation and Language

L2V-CoT: Cross-Modal Transfer of Chain-of-Thought Reasoning via Latent Intervention

Yuliang Zhan, Xinyu Tang, Han Wan et al. (+3 more)

Recently, Chain-of-Thought (CoT) reasoning has significantly enhanced the capabilities of large language models (LLMs), but Vision-Language Models (VLMs) still struggle with multi-step reasoning tasks due to limited multimodal reasoning data. To bridge this gap, researchers have explored methods to transfer CoT reasoning from LLMs to VLMs. However, existing approaches either need high training costs or require architectural alignment. In this paper, we use Linear Artificial Tomography (LAT) to empirically show that LLMs and VLMs share similar low-frequency latent representations of CoT reasoning despite architectural differences. Based on this insight, we propose L2V-CoT, a novel training-free latent intervention approach that transfers CoT reasoning from LLMs to VLMs. L2V-CoT extracts and resamples low-frequency CoT representations from LLMs in the frequency domain, enabling dimension matching and latent injection into VLMs during inference to enhance reasoning capabilities. Extensive experiments demonstrate that our approach consistently outperforms training-free baselines and even surpasses supervised methods.
Read the full abstract

All authors Yuliang Zhan, Xinyu Tang, Han Wan, Jian Li, Ji-Rong Wen, Hao Sun

Computation and Language

Principled Context Engineering for RAG: Statistical Guarantees via Conformal Prediction

Debashish Chakraborty, Eugene Yang, Daniel Khashabi et al. (+2 more)

Retrieval-Augmented Generation (RAG) enhances factual grounding in large language models (LLMs) by incorporating retrieved evidence, but LLM accuracy declines when long or noisy contexts exceed the model's effective attention span. Existing pre-generation filters rely on heuristics or uncalibrated LLM confidence scores, offering no statistical control over retained evidence. We evaluate and demonstrate context engineering through conformal prediction, a coverage-controlled filtering framework that removes irrelevant content while preserving recall of supporting evidence. Using both embedding- and LLM-based scoring functions, we test this approach on the NeuCLIR and RAGTIME collections. Conformal filtering consistently meets its target coverage, ensuring that a specified fraction of relevant snippets are retained, and reduces retained context by 2-3x relative to unfiltered retrieval. On NeuCLIR, downstream factual accuracy measured by ARGUE F1 improves under strict filtering and remains stable at moderate coverage, indicating that most discarded material is redundant or irrelevant. These results demonstrate that conformal prediction enables reliable, coverage-controlled context reduction in RAG, offering a model-agnostic and principled approach to context engineering.
Read the full abstract

All authors Debashish Chakraborty, Eugene Yang, Daniel Khashabi, Dawn Lawrie, Kevin Duh

Computation and Language · Cross-listed from cs.CV

When Better Teachers Don't Make Better Students: Revisiting Knowledge Distillation for CLIP Models in VQA

Pume Tuchinda, Parinthapat Pengpun, Romrawin Chumpu et al. (+3 more)

Vision-language models (VLMs) have achieved remarkable success across multimodal tasks, yet their substantial computational demands hinder efficient deployment. Knowledge distillation (KD) has emerged as a powerful approach for building lightweight but competitive models, with strong evidence from both language and vision domains. However, its application to VLMs, particularly CLIP-style models, remains limited, often constrained to small-scale teachers and narrow evaluation tasks such as classification or retrieval. In this work, we present the first systematic study of distillation across a range of CLIP-style teacher models, ranging from standard baselines to large-scale state-of-the-art models. Contrary to trends observed in NLP and vision, we find that stronger teachers do not consistently yield better students; in fact, existing distillation frameworks often fail to scale, leading to degraded performance in downstream multimodal tasks such as visual question answering. Our findings challenge prevailing assumptions in KD and point toward new directions for designing parameter-efficient multimodal models.
Read the full abstract

All authors Pume Tuchinda, Parinthapat Pengpun, Romrawin Chumpu, Patomporn Payoungkhamdee, Sarana Nutanong, Peerat Limkonchotiwat

Computation and Language

A superpersuasive autonomous policy debating system

Allen Roush, Devin Gonier, John Hines et al. (+4 more)

The capacity for highly complex, evidence-based, and strategically adaptive persuasion remains a formidable great challenge for artificial intelligence. Previous work, like IBM Project Debater, focused on generating persuasive speeches in simplified and shortened debate formats intended for relatively lay audiences. We introduce DeepDebater, a novel autonomous system capable of participating in and winning a full, unmodified, two-team competitive policy debate. Our system employs a hierarchical architecture of specialized multi-agent workflows, where teams of LLM-powered agents collaborate and critique one another to perform discrete argumentative tasks. Each workflow utilizes iterative retrieval, synthesis, and self-correction using a massive corpus of policy debate evidence (OpenDebateEvidence) and produces complete speech transcripts, cross-examinations, and rebuttals. We introduce a live, interactive end-to-end presentation pipeline that renders debates with AI speech and animation: transcripts are surface-realized and synthesized to audio with OpenAI TTS, and then displayed as talking-head portrait videos with EchoMimic V1. Beyond fully autonomous matches (AI vs AI), DeepDebater supports hybrid human-AI operation: human debaters can intervene at any stage, and humans can optionally serve as opponents against AI in any speech, allowing AI-human and AI-AI rounds. In preliminary evaluations against human-authored cases, DeepDebater produces qualitatively superior argumentative components and consistently wins simulated rounds as adjudicated by an independent autonomous judge. Expert human debate coaches also prefer the arguments, evidence, and cases constructed by DeepDebater. We open source all code, generated speech transcripts, audio and talking head video here: https://github.com/Hellisotherpeople/DeepDebater/tree/main
Read the full abstract

All authors Allen Roush, Devin Gonier, John Hines, Judah Goldfeder, Philippe Martin Wyder, Sanjay Basu, Ravid Shwartz Ziv

Friday 21 November

Computation and Language · Cross-listed from cs.IR

Enriching Taxonomies Using Large Language Models

Zeinab Ghamlouch and Mehwish Alam

Taxonomies play a vital role in structuring and categorizing information across domains. However, many existing taxonomies suffer from limited coverage and outdated or ambiguous nodes, reducing their effectiveness in knowledge retrieval. To address this, we present Taxoria, a novel taxonomy enrichment pipeline that leverages Large Language Models (LLMs) to enhance a given taxonomy. Unlike approaches that extract internal LLM taxonomies, Taxoria uses an existing taxonomy as a seed and prompts an LLM to propose candidate nodes for enrichment. These candidates are then validated to mitigate hallucinations and ensure semantic relevance before integration. The final output includes an enriched taxonomy with provenance tracking and visualization of the final merged taxonomy for analysis.
Read the full abstract

Computation and Language

EduMod-LLM: A Modular Approach for Designing Flexible and Transparent Educational Assistants

Meenakshi Mittal, Rishi Khare, Mihran Miroyan et al. (+2 more)

With the growing use of Large Language Model (LLM)-based Question-Answering (QA) systems in education, it is critical to evaluate their performance across individual pipeline components. In this work, we introduce {\model}, a modular function-calling LLM pipeline, and present a comprehensive evaluation along three key axes: function calling strategies, retrieval methods, and generative language models. Our framework enables fine-grained analysis by isolating and assessing each component. We benchmark function-calling performance across LLMs, compare our novel structure-aware retrieval method to vector-based and LLM-scoring baselines, and evaluate various LLMs for response synthesis. This modular approach reveals specific failure modes and performance patterns, supporting the development of interpretable and effective educational QA systems. Our findings demonstrate the value of modular function calling in improving system transparency and pedagogical alignment. Website and Supplementary Material: https://chancharikmitra.github.io/EduMod-LLM-website/
Read the full abstract

All authors Meenakshi Mittal, Rishi Khare, Mihran Miroyan, Chancharik Mitra, Narges Norouzi

Computation and Language

A Multiscale Geometric Method for Capturing Relational Topic Alignment

Conrad D. Hougen, Karl T. Pazdernik and Alfred O. Hero

Interpretable topic modeling is essential for tracking how research interests evolve within co-author communities. In scientific corpora, where novelty is prized, identifying underrepresented niche topics is particularly important. However, contemporary models built from dense transformer embeddings tend to miss rare topics and therefore also fail to capture smooth temporal alignment. We propose a geometric method that integrates multimodal text and co-author network data, using Hellinger distances and Ward's linkage to construct a hierarchical topic dendrogram. This approach captures both local and global structure, supporting multiscale learning across semantic and temporal dimensions. Our method effectively identifies rare-topic structure and visualizes smooth topic drift over time. Experiments highlight the strength of interpretable bag-of-words models when paired with principled geometric alignment.
Read the full abstract

Computation and Language

A cross-species neural foundation model for end-to-end speech decoding

Yizi Zhang, Linyang He, Chaofei Fan et al. (+9 more)

Speech brain-computer interfaces (BCIs) aim to restore communication for people with paralysis by translating neural activity into text. Most systems use cascaded frameworks that decode phonemes before assembling sentences with an n-gram language model (LM), preventing joint optimization of all stages simultaneously. Here, we introduce an end-to-end BraIn-to-Text (BIT) framework that translates neural activity into coherent sentences using a single differentiable neural network. Central to our approach is a cross-task, cross-species pretrained neural encoder, whose representations transfer to both attempted and imagined speech. In a cascaded setting with an n-gram LM, the pretrained encoder establishes a new state-of-the-art (SOTA) on the Brain-to-Text '24 and '25 benchmarks. Integrated end-to-end with audio large language models (LLMs) and trained with contrastive learning for cross-modal alignment, BIT reduces the word error rate (WER) of the prior end-to-end method from 24.69% to 10.22%. Notably, we find that small-scale audio LLMs markedly improve end-to-end decoding. Beyond record-setting performance, BIT aligns attempted and imagined speech embeddings to enable cross-task generalization. Altogether, our approach advances the integration of large, diverse neural datasets, paving the way for an end-to-end decoding framework that supports seamless, differentiable optimization.
Read the full abstract

All authors Yizi Zhang, Linyang He, Chaofei Fan, Tingkai Liu, Han Yu, Trung Le, Jingyuan Li, Scott Linderman, Lea Duncker, Francis R Willett, Nima Mesgarani, Liam Paninski

Computation and Language

Polarity-Aware Probing for Quantifying Latent Alignment in Language Models

Sabrina Sadiekh, Elena Ericheva and Chirag Agarwal

Advances in unsupervised probes such as Contrast-Consistent Search (CCS), which reveal latent beliefs without relying on token outputs, raise the question of whether these methods can reliably assess model alignment. We investigate this by examining the sensitivity of CCS to harmful vs. safe statements and by introducing Polarity-Aware CCS (PA-CCS), a method for evaluating whether a model's internal representations remain consistent under polarity inversion. We propose two alignment-oriented metrics, Polar-Consistency and the Contradiction Index, to quantify the semantic robustness of a model's latent knowledge. To validate PA-CCS, we curate two main datasets and one control dataset containing matched harmful-safe sentence pairs constructed using different methodologies (concurrent and antagonistic statements). We apply PA-CCS to 16 language models. Our results show that PA-CCS identifies both architectural and layer-specific differences in the encoding of latent harmful knowledge. Notably, replacing the negation token with a meaningless marker degrades PA-CCS scores for models with well-aligned internal representations, while models lacking robust internal calibration do not exhibit this degradation. Our findings highlight the potential of unsupervised probing for alignment evaluation and emphasize the need to incorporate structural robustness checks into interpretability benchmarks. Code and datasets are available at: https://github.com/SadSabrina/polarity-probing. WARNING: This paper contains potentially sensitive, harmful, and offensive content.
Read the full abstract

Computation and Language

R2Q: Towards Robust 2-Bit Large Language Models via Residual Refinement Quantization

Jiayi Chen, Jieqi Shi, Jing Huo et al. (+1 more)

The rapid progress of Large Language Models (LLMs) has brought substantial computational and memory demands, spurring the adoption of low-bit quantization. While 8-bit and 4-bit formats have become prevalent, extending quantization to 2 bits remains challenging due to severe accuracy degradation. To address this, we propose Residual Refinement Quantization (R2Q)-a novel 2-bit quantization framework that decomposes the process into two sequential 1-bit sub-quantizations, forming an adaptive quantization lattice. Extensive evaluations on Llama, OPT, and Qwen across diverse benchmarks-covering question answering, commonsense reasoning, and language modeling-demonstrate that R2Q consistently outperforms existing 2-bit quantization methods in both fine-grained and coarse-grained settings. By refining quantization through a residual learning mechanism, R2Q enhances performance, improves training stability, and accelerates convergence under extreme compression. Furthermore, its modular design enables seamless integration with existing quantization-aware training (QAT) frameworks.
Read the full abstract

All authors Jiayi Chen, Jieqi Shi, Jing Huo, Chen Wu

Computation and Language

Closing the Performance Gap Between AI and Radiologists in Chest X-Ray Reporting

Harshita Sharma, Maxwell C. Reynolds, Valentina Salvatelli et al. (+26 more)

AI-assisted report generation offers the opportunity to reduce radiologists' workload stemming from expanded screening guidelines, complex cases and workforce shortages, while maintaining diagnostic accuracy. In addition to describing pathological findings in chest X-ray reports, interpreting lines and tubes (L&T) is demanding and repetitive for radiologists, especially with high patient volumes. We introduce MAIRA-X, a clinically evaluated multimodal AI model for longitudinal chest X-ray (CXR) report generation, that encompasses both clinical findings and L&T reporting. Developed using a large-scale, multi-site, longitudinal dataset of 3.1 million studies (comprising 6 million images from 806k patients) from Mayo Clinic, MAIRA-X was evaluated on three holdout datasets and the public MIMIC-CXR dataset, where it significantly improved AI-generated reports over the state of the art on lexical quality, clinical correctness, and L&T-related elements. A novel L&T-specific metrics framework was developed to assess accuracy in reporting attributes such as type, longitudinal change and placement. A first-of-its-kind retrospective user evaluation study was conducted with nine radiologists of varying experience, who blindly reviewed 600 studies from distinct subjects. The user study found comparable rates of critical errors (3.0% for original vs. 4.6% for AI-generated reports) and a similar rate of acceptable sentences (97.8% for original vs. 97.4% for AI-generated reports), marking a significant improvement over prior user studies with larger gaps and higher error rates. Our results suggest that MAIRA-X can effectively assist radiologists, particularly in high-volume clinical settings.
Read the full abstract

All authors Harshita Sharma, Maxwell C. Reynolds, Valentina Salvatelli, Anne-Marie G. Sykes, Kelly K. Horst, Anton Schwaighofer, Maximilian Ilse, Olesya Melnichenko, Sam Bond-Taylor, Fernando Pérez-García, Vamshi K. Mugu, Alex Chan, Ceylan Colak, Shelby A. Swartz, Motassem B. Nashawaty, Austin J. Gonzalez, Heather A. Ouellette, Selnur B. Erdal, Beth A. Schueler, Maria T. Wetscherek, Noel Codella, Mohit Jain, Shruthi Bannur, Kenza Bouzid, Daniel C. Castro, Stephanie Hyland, Panos Korfiatis, Ashish Khandelwal, Javier Alvarez-Valle

Computation and Language

Asking LLMs to Verify First is Almost Free Lunch

Shiguang Wu and Quanming Yao

To enhance the reasoning capabilities of Large Language Models (LLMs) without high costs of training, nor extensive test-time sampling, we introduce Verification-First (VF), a strategy that prompts models to verify a provided candidate answer, even a trivial or random one, before generating a solution. This approach triggers a "reverse reasoning" process complementary to standard forward Chain-of-Thought (CoT), which restricts the logical search space of the answer by pruning the LLM's output distribution. We further generalize VF prompting to Iter-VF, a sequential test-time scaling (TTS) method that iteratively cycles the verification-generation process using the model's previous answer. Extensive experiments across various benchmarks and various LLMs confirm that VF prompting with random answer consistently outperforms standard CoT with minimal computational overhead, and Iter-VF outperforms existing TTS strategies. VF is also effective on SOTA thinking models. For example, by using the simple VF prompting, we obtain a new SOTA 94.9% accuracy on GPQA-Diamond with Gemini-3-Pro-Preview where VF reduces its errors by ~30% relatively.
Read the full abstract

Computation and Language

RoSA: Enhancing Parameter-Efficient Fine-Tuning via RoPE-aware Selective Adaptation in Large Language Models

Dayan Pan, Jingyuan Wang, Yilong Zhou et al. (+3 more)

Fine-tuning large language models is essential for task-specific adaptation, yet it remains computationally prohibitive. Parameter-Efficient Fine-Tuning (PEFT) methods have emerged as a solution, but current approaches typically ignore the distinct roles of model components and the heterogeneous importance across layers, thereby limiting adaptation efficiency. Motivated by the observation that Rotary Position Embeddings (RoPE) induce critical activations in the low-frequency dimensions of attention states, we propose RoPE-aware Selective Adaptation (RoSA), a novel PEFT framework that allocates trainable parameters in a more targeted and effective manner. RoSA comprises a RoPE-aware Attention Enhancement (RoAE) module, which selectively enhances the low-frequency components of RoPE-influenced attention states, and a Dynamic Layer Selection (DLS) strategy that adaptively identifies and updates the most critical layers based on LayerNorm gradient norms. By combining dimension-wise enhancement with layer-wise adaptation, RoSA achieves more targeted and efficient fine-tuning. Extensive experiments on fifteen commonsense and arithmetic benchmarks demonstrate that RoSA outperforms existing mainstream PEFT methods under comparable trainable parameters. The code is available to ease reproducibility at https://github.com/Applied-Machine-Learning-Lab/RoSA.
Read the full abstract

All authors Dayan Pan, Jingyuan Wang, Yilong Zhou, Jiawei Cheng, Pengyue Jia, Xiangyu Zhao

Computation and Language

HUMORCHAIN: Theory-Guided Multi-Stage Reasoning for Interpretable Multimodal Humor Generation

Jiajun Zhang, Shijia Luo, Ruikang Zhang et al. (+1 more)

Humor, as both a creative human activity and a social binding mechanism, has long posed a major challenge for AI generation. Although producing humor requires complex cognitive reasoning and social understanding, theories of humor suggest that it follows learnable patterns and structures, making it theoretically possible for generative models to acquire them implicitly. In recent years, multimodal humor has become a prevalent form of online communication, especially among Gen Z, highlighting the need for AI systems capable of integrating visual understanding with humorous language generation. However, existing data-driven approaches lack explicit modeling or theoretical grounding of humor, often producing literal descriptions that fail to capture its underlying cognitive mechanisms, resulting in the generated image descriptions that are fluent but lack genuine humor or cognitive depth. To address this limitation, we propose HUMORCHAIN (HUmor-guided Multi-step Orchestrated Reasoning Chain for Image Captioning), a theory-guided multi-stage reasoning framework. It integrates visual semantic parsing, humor- and psychology-based reasoning, and a fine-tuned discriminator for humor evaluation, forming an interpretable and controllable cognitive reasoning chain. To the best of our knowledge, this is the first work to explicitly embed cognitive structures from humor theories into multimodal humor generation, enabling a structured reasoning process from visual understanding to humor creation. Experiments on Meme-Image-No-Text, Oogiri-GO, and OxfordTVG-HIC datasets show that HUMORCHAIN outperforms state-of-the-art baselines in human humor preference, Elo/BT scores, and semantic diversity, demonstrating that theory-driven structured reasoning enables large language models to generate humor aligned with human perception.
Read the full abstract

All authors Jiajun Zhang, Shijia Luo, Ruikang Zhang, Qi Su

Computation and Language

Identifying Quantum Structure in AI Language: Evidence for Evolutionary Convergence of Human and Artificial Cognition

Diederik Aerts, Jonito Aerts Arguëlles, Lester Beltran et al. (+4 more)

We present the results of cognitive tests on conceptual combinations, performed using specific Large Language Models (LLMs) as test subjects. In the first test, performed with ChatGPT and Gemini, we show that Bell's inequalities are significantly violated, which indicates the presence of a 'non-classical probability model' with probabilities that do not satisfy Kolmogorov's axioms. In the second test, also performed using ChatGPT and Gemini, we identify the presence of 'Bose-Einstein statistics', rather than the intuitively expected 'Maxwell-Boltzmann statistics', in the distribution of the words contained in large-size texts. Interestingly, these findings mirror the results previously obtained in both cognitive tests with human participants and information retrieval tests on large corpora. Taken together, they point to the 'systematic emergence of non-classical quantum-like structures in conceptual-linguistic domains', regardless of whether the cognitive agent is human or artificial. Although LLMs are classified as neural networks for historical reasons, we believe that a more essential form of knowledge organization takes place in the distributive semantic structure of vector spaces built on top of the neural network. It is this meaning-bearing structure that lends itself to a phenomenon of evolutionary convergence between human cognition and language, slowly established through biological evolution, and LLM cognition and language, emerging much more rapidly as a result of self-learning and training. We analyze various aspects and examples that contain evidence supporting the above hypothesis. We also advance a unifying framework that explains the pervasive quantum organization of meaning that we identify.
Read the full abstract

All authors Diederik Aerts, Jonito Aerts Arguëlles, Lester Beltran, Suzette Geriente, Roberto Leporini, Massimiliano Sassoli de Bianchi, Sandro Sozzo

Computation and Language

A Benchmark for Procedural Memory Retrieval in Language Agents

Ishant Kohar and Aswanth Krishnan

Current AI agents excel in familiar settings, but fail sharply when faced with novel tasks with unseen vocabularies – a core limitation of procedural memory systems. We present the first benchmark that isolates procedural memory retrieval from task execution, evaluating whether agents can recognize functionally equivalent procedures that span different object instantiations. Using ALFWorld, we construct dual corpora of expert and LLM-generated trajectories and evaluate six retrieval methods using systematically stratified queries. Our results expose a clear generalization cliff: embedding-based methods perform strongly on familiar contexts, yet degrade considerably on novel ones, while LLM-generated procedural abstractions demonstrate reliable cross-context transfer. Controlled ablations show that although embeddings capture some lexical-level abstraction, they fundamentally treat procedures as unordered bags of words, discarding temporal structure necessary for cross-context transfer. Corpus scale delivers far larger gains than representation enrichment, revealing an architectural ceiling in current encoders. Our benchmark offers the first diagnostic framework separating genuine procedural understanding from surface-level memorization and gives tools for developing retrieval systems capable of dependable generalization. Resources available at our GitHub repository (https://github.com/qpiai/Proced_mem_bench).
Read the full abstract

Computation and Language

Beyond Component Strength: Synergistic Integration and Adaptive Calibration in Multi-Agent RAG Systems

Jithin Krishnan

Building reliable retrieval-augmented generation (RAG) systems requires more than adding powerful components; it requires understanding how they interact. Using ablation studies on 50 queries (15 answerable, 10 edge cases, and 25 adversarial), we show that enhancements such as hybrid retrieval, ensemble verification, and adaptive thresholding provide almost no benefit when used in isolation, yet together achieve a 95% reduction in abstention (from 40% to 2%) without increasing hallucinations. We also identify a measurement challenge: different verification strategies can behave safely but assign inconsistent labels (for example, "abstained" versus "unsupported"), creating apparent hallucination rates that are actually artifacts of labeling. Our results show that synergistic integration matters more than the strength of any single component, that standardized metrics and labels are essential for correctly interpreting performance, and that adaptive calibration is needed to prevent overconfident over-answering even when retrieval quality is high.
Read the full abstract

Computation and Language

Affective Multimodal Agents with Proactive Knowledge Grounding for Emotionally Aligned Marketing Dialogue

Lin Yu, Xiaofei Han, Yifei Kang et al. (+4 more)

Recent advances in large language models (LLMs) have enabled fluent dialogue systems, but most remain reactive and struggle in emotionally rich, goal-oriented settings such as marketing conversations. To address this limitation, we propose AffectMind, a multimodal affective dialogue agent that performs proactive reasoning and dynamic knowledge grounding to sustain emotionally aligned and persuasive interactions. AffectMind combines three components: a Proactive Knowledge Grounding Network (PKGN) that continuously updates factual and affective context from text, vision, and prosody; an Emotion–Intent Alignment Model (EIAM) that jointly models user emotion and purchase intent to adapt persuasion strategies; and a Reinforced Discourse Loop (RDL) that optimizes emotional coherence and engagement via reinforcement signals from user responses. Experiments on two newly curated marketing dialogue datasets, MM-ConvMarket and AffectPromo, show that AffectMind outperforms strong LLM-based baselines in emotional consistency (+26%), persuasive success rate (+19%), and long-term user engagement (+23%), highlighting emotion-grounded proactivity as a key capability for commercial multimodal agents.
Read the full abstract

All authors Lin Yu, Xiaofei Han, Yifei Kang, Chiung-Yi Tseng, Danyang Zhang, Ziqian Bi, Zhimo Han

Computation and Language · Cross-listed from cs.LG

Deterministic Inference across Tensor Parallel Sizes That Eliminates Training-Inference Mismatch

Ziyang Zhang, Xinheng Ding, Jiayi Yuan et al. (+4 more)

Deterministic inference is increasingly critical for large language model (LLM) applications such as LLM-as-a-judge evaluation, multi-agent systems, and Reinforcement Learning (RL). However, existing LLM serving frameworks exhibit non-deterministic behavior: identical inputs can yield different outputs when system configurations (e.g., tensor parallel (TP) size, batch size) vary, even under greedy decoding. This arises from the non-associativity of floating-point arithmetic and inconsistent reduction orders across GPUs. While prior work has addressed batch-size-related nondeterminism through batch-invariant kernels, determinism across different TP sizes remains an open problem, particularly in RL settings, where the training engine typically uses Fully Sharded Data Parallel (i.e., TP = 1) while the rollout engine relies on multi-GPU TP to maximize the inference throughput, creating a natural mismatch between the two. This precision mismatch problem may lead to suboptimal performance or even collapse for RL training. We identify and analyze the root causes of TP-induced inconsistency and propose Tree-Based Invariant Kernels (TBIK), a set of TP-invariant matrix multiplication and reduction primitives that guarantee bit-wise identical results regardless of TP size. Our key insight is to align intra- and inter-GPU reduction orders through a unified hierarchical binary tree structure. We implement these kernels in Triton and integrate them into vLLM and FSDP. Experiments confirm zero probability divergence and bit-wise reproducibility for deterministic inference across different TP sizes. Also, we achieve bit-wise identical results between vLLM and FSDP in RL training pipelines with different parallel strategy. Code is available at https://github.com/nanomaoli/llm_reproducibility.
Read the full abstract

All authors Ziyang Zhang, Xinheng Ding, Jiayi Yuan, Rixin Liu, Huizi Mao, Jiarong Xing, Zirui Liu

Computation and Language

Point of Order: Action-Aware LLM Persona Modeling for Data-Grounded Civic Deliberation

Scott Merrill and Shashank Srivastava

LLM-based simulations can enable controlled studies of civic deliberation, but current systems lack speaker-attributed data and methods for evaluating long-form institutional behavior. ASR transcripts typically use anonymous labels such as $Speaker\_1$, preventing models from learning stable participant behavior across meetings. We present a reproducible pipeline that converts public Zoom recordings into speaker-attributed transcripts enriched with persona profiles, topics, and pragmatic "action tags" such as $[propose\_motion]$. Using this pipeline, we release three public datasets of government deliberation (Appellate Court hearings, School Board meetings, and Municipal Council sessions) and fine-tune LLM personas on this action-aware data. We evaluate simulations along four dimensions: persona fidelity, persona consistency, institutional fidelity, and behavioral coherence. Action-aware fine-tuning cuts perplexity by 67%, doubles classifier-based persona fidelity, increases vote attempts by up to $3.6\times$, and improves deliberative responsiveness by up to 70%. Human evaluations show that simulated excerpts are often hard to distinguish from real deliberations, indicating a practical foundation for data-grounded civic simulation studies.
Read the full abstract

Computation and Language

PoETa v2: Toward More Robust Evaluation of Large Language Models in Portuguese

Thales Sales Almeida, Ramon Pires, Hugo Abonizio et al. (+2 more)

Large Language Models (LLMs) exhibit significant variations in performance across linguistic and cultural contexts, underscoring the need for systematic evaluation in diverse languages. In this work, we present the most extensive evaluation of LLMs for the Portuguese language to date. Leveraging our newly introduced PoETa v2 benchmark – a comprehensive suite of over 40 tasks in Portuguese – we assess more than 20 models covering a broad spectrum of training scales and computational resources. Our study reveals how computational investment and language-specific adaptation impact performance in Portuguese, while also analyzing performance gaps in comparison to equivalent tasks in English. Through this benchmark and analysis, PoETa v2 lays the groundwork for future research on Portuguese language modeling and evaluation. The benchmark is available at https://github.com/PoETaV2/PoETaV2.
Read the full abstract

All authors Thales Sales Almeida, Ramon Pires, Hugo Abonizio, Rodrigo Nogueira, Hélio Pedrini

Computation and Language

Computational frame analysis revisited: On LLMs for studying news coverage

Sharaj Kunjar, Alyssa Hasegawa Smith, Tyler R Mckenzie et al. (+3 more)

Computational approaches have previously shown various promises and pitfalls when it comes to the reliable identification of media frames. Generative LLMs like GPT and Claude are increasingly being used as content analytical tools, but how effective are they for frame analysis? We address this question by systematically evaluating them against their computational predecessors: bag-of-words models and encoder-only transformers; and traditional manual coding procedures. Our analysis rests on a novel gold standard dataset that we inductively and iteratively developed through the study, investigating six months of news coverage of the US Mpox epidemic of 2022. While we discover some potential applications for generative LLMs, we demonstrate that they were consistently outperformed by manual coders, and in some instances, by smaller language models. Some form of human validation was always necessary to determine appropriate model choice. Additionally, by examining how the suitability of various approaches depended on the nature of different tasks that were part of our frame analytical workflow, we provide insights as to how researchers may leverage the complementarity of these approaches to use them in tandem. We conclude by endorsing a methodologically pluralistic approach and put forth a roadmap for computational frame analysis for researchers going forward.
Read the full abstract

All authors Sharaj Kunjar, Alyssa Hasegawa Smith, Tyler R Mckenzie, Rushali Mohbe, Samuel V Scarpino, Brooke Foucault Welles

Computation and Language · Cross-listed from cs.LG

DeepCoT: Deep Continual Transformers for Real-Time Inference on Data Streams

Ginés Carreto Picón, Peng Yuan Zhou, Qi Zhang et al. (+1 more)

Transformer-based models have dramatically increased their size and parameter count to tackle increasingly complex tasks. At the same time, there is a growing demand for high performance, low-latency inference on devices with limited resources. In particular, stream data inference is typically performed over a sliding temporal window, leading to highly redundant computations. While the recent Continual Transformers started addressing this issue, they can be effectively used only in shallow models, which limits their scope and generalization power. In this paper, we propose the Deep Continual Transformer (DeepCoT), a redundancy-free encoder attention mechanism that can be applied over existing deep encoder architectures with minimal changes. In our experiments over audio, video, and text streams, we show that DeepCoTs retain comparative performance to their non-continual baselines while offering a linear computational cost for all Transformer layers, which reduces up to two orders of magnitude in the running time compared to previous efficient models.
Read the full abstract

All authors Ginés Carreto Picón, Peng Yuan Zhou, Qi Zhang, Alexandros Iosifidis

Computation and Language · Cross-listed from cs.CY

A Cross-Cultural Assessment of Human Ability to Detect LLM-Generated Fake News about South Africa

Tim Schlippe, Matthias Wölfel and Koena Ronny Mabokela

This study investigates how cultural proximity affects the ability to detect AI-generated fake news by comparing South African participants with those from other nationalities. As large language models increasingly enable the creation of sophisticated fake news, understanding human detection capabilities becomes crucial, particularly across different cultural contexts. We conducted a survey where 89 participants (56 South Africans, 33 from other nationalities) evaluated 10 true South African news articles and 10 AI-generated fake versions. Results reveal an asymmetric pattern: South Africans demonstrated superior performance in detecting true news about their country (40% deviation from ideal rating) compared to other participants (52%), but performed worse at identifying fake news (62% vs. 55%). This difference may reflect South Africans' higher overall trust in news sources. Our analysis further shows that South Africans relied more on content knowledge and contextual understanding when judging credibility, while participants from other countries emphasised formal linguistic features such as grammar and structure. Overall, the deviation from ideal rating was similar between groups (51% vs. 53%), suggesting that cultural familiarity appears to aid verification of authentic information but may also introduce bias when evaluating fabricated content. These insights contribute to understanding cross-cultural dimensions of misinformation detection and inform strategies for combating AI-generated fake news in increasingly globalised information ecosystems where content crosses cultural and geographical boundaries.
Read the full abstract

Computation and Language · Cross-listed from cs.DB

LLM and Agent-Driven Data Analysis: A Systematic Approach for Enterprise Applications and System-level Deployment

Xi Wang, Xianyao Ling, Kun Li et al. (+5 more)

The rapid progress in Generative AI and Agent technologies is profoundly transforming enterprise data management and analytics. Traditional database applications and system deployment are fundamentally impacted by AI-driven tools, such as Retrieval-Augmented Generation (RAG) and vector database technologies, which provide new pathways for semantic querying over enterprise knowledge bases. In the meantime, data security and compliance are top priorities for organizations adopting AI technologies. For enterprise data analysis, SQL generations powered by large language models (LLMs) and AI agents, has emerged as a key bridge connecting natural language with structured data, effectively lowering the barrier to enterprise data access and improving analytical efficiency. This paper focuses on enterprise data analysis applications and system deployment, covering a range of innovative frameworks, enabling complex query understanding, multi-agent collaboration, security verification, and computational efficiency. Through representative use cases, key challenges related to distributed deployment, data security, and inherent difficulties in SQL generation tasks are discussed.
Read the full abstract

All authors Xi Wang, Xianyao Ling, Kun Li, Gang Yin, Liang Zhang, Jiang Wu, Annie Wang, Weizhe Wang

Computation and Language · Cross-listed from cs.AI

Bridging Symbolic Control and Neural Reasoning in LLM Agents – The Structured Cognitive Loop

Myung Ho Kim

Large language model agents suffer from architectural fragilities such as entangled reasoning and execution, memory volatility, and uncontrolled action sequences. We introduce Structured Cognitive Loop (SCL), a modular agent architecture that separates cognition into Retrieval, Cognition, Control, Action, and Memory (R-CCAM). SCL introduces Regulation as a dedicated governance layer through which Soft Symbolic Control applies symbolic constraints to probabilistic inference, while Control remains a distinct deterministic runtime engine for duplicate-call prevention, error limits, and termination judgment. Through multi-step conditional reasoning experiments, we show that SCL achieves zero policy violations, prevents redundant tool calls, and maintains complete decision traceability. We position SCL within hybrid intelligence, distinguish it from prompt-centric, memory-only, and neuro-symbolic approaches, and derive three design principles for trustworthy agents: modular decomposition, adaptive symbolic governance, and transparent state management. With an open-source implementation and a live GPT-4o-powered travel planning agent, this work offers a practical path toward reliable, explainable, and governable LLM agents.
Read the full abstract

Computation and Language · Cross-listed from cs.CR

Cross-User Poisoning: User-Task Boundary Failures in Multi-User Collaborative Language Agents

Atharv Singh Patlan, Peiyao Sheng, S. Ashwin Hebbar et al. (+2 more)

Language agents are moving from single-user assistants to shared collaborators in workspaces, forums, and group chats. A shared agent observes interleaved messages from multiple users, maintains persistent context, and executes tool actions, creating a boundary problem absent from standard single-user agents: the agent must decide not only whether an instruction is safe, but which user or task it is allowed to govern. We identify cross-user poisoning (CUP), where an adversary injects a message into shared context that is later applied while the agent serves a benign user, causing unauthorized actions or responses outside the instruction's intended scope. We validate CUP on two deployed multi-user agents, Continua and ElizaOS, and introduce MURMUR, a framework for evaluating shared-context agents under concurrent multi-user interactions. Across Slack, Workspace, and Airline domains, CUP achieves high attack success, persists across later interactions, and remains substantially more effective than matched prompt-injection attacks. We evaluate boundary-scoping defenses and find that context summarization alone is insufficient, while task clustering and provenance prompting substantially reduce non-adaptive propagation. These results show that robust multi-user agents require explicit user/task scoping rather than generic input filtering or context compression.
Read the full abstract

All authors Atharv Singh Patlan, Peiyao Sheng, S. Ashwin Hebbar, Prateek Mittal, Pramod Viswanath

Computation and Language

Masked-and-Reordered Self-Supervision for Reinforcement Learning from Verifiable Rewards

Zhen Wang, Zhifeng Gao and Guolin Ke

Test-time scaling has been shown to substantially improve large language models' (LLMs) mathematical reasoning. However, for a large portion of mathematical corpora, especially theorem proving, RLVR's scalability is limited: intermediate reasoning is crucial, while final answers are difficult to directly and reliably verify. Meanwhile, token-level SFT often degenerates into rote memorization rather than inducing longer chains of thought. Inspired by BERT's self-supervised tasks, we propose MR-RLVR (Masked-and-Reordered RLVR), which constructs process-level self-supervised rewards via "masked-then-fill" and "step reordering" to extract learnable signals from intermediate reasoning. Our training pipeline comprises two stages: we first perform self-supervised training on sampled mathematical calculation and proof data; we then conduct RLVR fine-tuning on mathematical calculation datasets where only outcomes are verifiable. We implement MR-RLVR on Qwen2.5-3B and DeepSeek-R1-Distill-Qwen-1.5B, and evaluate on AIME24, AIME25, AMC23, and MATH500. Under a fixed sampling and decoding budget, MR-RLVR achieves average relative gains over the original RLVR of +9.86% Pass@1, +5.27% Pass@5, and +4.00% Pass@8. These results indicate that incorporating process-aware self-supervised signals can effectively enhance RLVR's scalability and performance in only outcome-verifiable settings.
Read the full abstract