Computation and Language
Mosh Levy, Zohar Elyoseph, Shauli Ravfogel et al. (+1 more)
Large
Language Models (LLMs) can generate
reasoning tokens before their final answer to boost performance on complex tasks. While these sequences seem like human thought processes, empirical evidence reveals that they are not a faithful explanation of the model's actual
reasoning process. To address this gap between appearance and function, we introduce the State over
Tokens (SoT) conceptual framework. SoT reframes
reasoning tokens not as a linguistic narrative, but as an externalized computational state – the sole persistent information carrier across the model's stateless generation cycles. This explains how the
tokens can drive correct
reasoning without being a faithful explanation when read as text and surfaces previously overlooked research questions on these
tokens. We argue that to truly understand the process that LLMs do, research must move beyond reading the
reasoning tokens as text and focus on decoding them as state.
Read the full abstract ▾
All authors Mosh Levy, Zohar Elyoseph, Shauli Ravfogel, Yoav Goldberg
Computation and Language
Persistent Personas? Role-Playing, Instruction Following, and Safety in Extended Interactions
Pedro Henrique Luz de Araujo, Michael A. Hedderich, Ali Modarressi et al. (+2 more)
Persona-assigned large
language models (LLMs) are used in domains such as education, healthcare, and sociodemographic simulation. Yet, they are typically evaluated only in short, single-round settings that do not reflect real-world usage. We introduce an evaluation protocol that combines long persona dialogues (over 100 rounds) and evaluation
datasets to create dialogue-conditioned
benchmarks that can robustly measure long-context effects. We then investigate the effects of dialogue length on persona fidelity, instruction-following, and safety of seven state-of-the-art open- and closed-weight LLMs. We find that persona fidelity degrades over the course of dialogues, especially in goal-oriented conversations, where models must sustain both persona fidelity and instruction following. We identify a trade-off between fidelity and instruction following, with non-persona baselines initially outperforming persona-assigned models; as dialogues progress and fidelity fades, persona responses become increasingly similar to baseline responses. Our findings highlight the fragility of persona applications in extended interactions and our work provides a protocol to systematically measure such failures.
Read the full abstract ▾
All authors Pedro Henrique Luz de Araujo, Michael A. Hedderich, Ali Modarressi, Hinrich Schuetze, Benjamin Roth
Computation and Language
Curió-Edu 7B: Examining Data Selection Impacts in LLM Continued Pretraining
Thales Sales Almeida, Rodrigo Nogueira and Hélio Pedrini
Continued pretraining extends a
language model's capabilities by further exposing it to additional data, often tailored to a specific linguistic or domain context. This strategy has emerged as an efficient alternative to full retraining when adapting general-purpose models to new settings. In this work, we investigate this paradigm through Curió 7B, a 7-billion-parameter model derived from LLaMA-2 and trained on 100 billion Portuguese
tokens from the ClassiCC-PT corpus - the most extensive Portuguese-specific continued-pretraining effort above the three-billion-parameter scale to date. Beyond scale, we investigate whether quantity alone suffices or whether data quality plays a decisive role in linguistic adaptation. To this end, we introduce Curió-Edu 7B, a variant trained exclusively on the educational and STEM-filtered subset of the same corpus, totaling just 10 billion
tokens. Despite using only 10% of the data and 20% of the computation, Curió-Edu 7B surpasses the full-corpus model in our evaluations, demonstrating that data selection can be fundamental even when adapting models with limited prior exposure to the target
language. The developed models are available at https://huggingface.co/collections/ClassiCC-Corpus/curio-edu
Read the full abstract ▾
Computation and Language · Cross-listed from cs.IR
Intelligent Scientific Literature Explorer using Machine Learning (ISLE)
Sina Jani, Arman Heidari, Amirmohammad Anvari et al. (+1 more)
The rapid acceleration of scientific publishing has created substantial challenges for researchers attempting to discover, contextualize, and interpret relevant literature. Traditional keyword-based search systems provide limited semantic understanding, while existing AI-driven tools typically focus on isolated tasks such as
retrieval, clustering, or bibliometric visualization. This paper presents an integrated system for scientific literature exploration that combines large-scale data acquisition, hybrid
retrieval, semantic topic modeling, and heterogeneous knowledge graph construction. The system builds a comprehensive corpus by merging full-text data from arXiv with structured metadata from OpenAlex. A hybrid
retrieval architecture fuses BM25 lexical search with
embedding-based semantic search using Reciprocal Rank Fusion. Topic modeling is performed on retrieved results using BERTopic or non-negative
matrix factorization depending on computational resources. A knowledge graph unifies papers, authors, institutions, countries, and extracted topics into an interpretable structure. The system provides a multi-layered exploration environment that reveals not only relevant publications but also the conceptual and relational landscape surrounding a query. Evaluation across multiple queries demonstrates improvements in
retrieval relevance, topic coherence, and interpretability. The proposed framework contributes an extensible foundation for AI-assisted scientific discovery.
Read the full abstract ▾
All authors Sina Jani, Arman Heidari, Amirmohammad Anvari, Zahra Rahimi
Computation and Language
NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agents
Jingzhe Ding, Shengda Long, Changxin Pu et al. (+46 more)
Recent advances in coding
agents suggest rapid progress toward autonomous software development, yet existing
benchmarks fail to rigorously evaluate the long-horizon capabilities required to build complete software systems. Most prior evaluations focus on localized code generation, scaffolded completion, or short-term repair tasks, leaving open the question of whether
agents can sustain coherent
reasoning, planning, and execution over the extended horizons demanded by real-world repository construction. To address this gap, we present NL2Repo Bench, a
benchmark explicitly designed to evaluate the long-horizon repository generation ability of coding
agents. Given only a single natural-
language requirements document and an empty workspace,
agents must autonomously design the architecture, manage dependencies, implement multi-module logic, and produce a fully installable Python library. Our experiments across state-of-the-art open- and closed-source models reveal that long-horizon repository generation remains largely unsolved: even the strongest
agents achieve below 40% average test pass rates and rarely complete an entire repository correctly. Detailed analysis uncovers fundamental long-horizon failure modes, including premature termination, loss of global coherence, fragile cross-file dependencies, and inadequate planning over hundreds of interaction steps. NL2Repo Bench establishes a rigorous, verifiable testbed for measuring sustained agentic competence and highlights long-horizon
reasoning as a central bottleneck for the next generation of autonomous coding
agents.
Read the full abstract ▾
All authors Jingzhe Ding, Shengda Long, Changxin Pu, Huan Zhou, Hongwan Gao, Xiang Gao, Chao He, Yue Hou, Fei Hu, Zhaojian Li, Weiran Shi, Zaiyuan Wang, Daoguang Zan, Chenchen Zhang, Xiaoxu Zhang, Qizhi Chen, Xianfu Cheng, Bo Deng, Qingshui Gu, Kai Hua, Juntao Lin, Pai Liu, Mingchen Li, Xuanguang Pan, Zifan Peng, Yujia Qin, Yong Shan, Zhewen Tan, Weihao Xie, Zihan Wang, Yishuo Yuan, Jiayu Zhang, Enduo Zhao, Yunfei Zhao, He Zhu, Liya Zhu, Chenyang Zou, Ming Ding, Jianpeng Jiao, Jiaheng Liu, Minghao Liu, Qian Liu, Chongyang Tao, Jian Yang, Tong Yang, Zhaoxiang Zhang, Xinjie Chen, Wenhao Huang, Ge Zhang
Computation and Language
CoDA: A Context-Decoupled Hierarchical Agent with Reinforcement Learning
Xuanzhang Liu, Jianglun Feng, Zhuoran Zhuang et al. (+7 more)
Large
Language Model (LLM)
agents trained with reinforcement learning (RL) show great promise for solving complex, multi-step tasks. However, their performance is often crippled by "Context Explosion", where the accumulation of long text outputs overwhelms the model's context window and leads to
reasoning failures. To address this, we introduce CoDA, a Context-Decoupled hierarchical
Agent, a simple but effective reinforcement learning framework that decouples high-level planning from low-level execution. It employs a single, shared LLM backbone that learns to operate in two distinct, contextually isolated roles: a high-level Planner that decomposes tasks within a concise strategic context, and a low-level Executor that handles tool interactions in an ephemeral, isolated workspace. We train this unified
agent end-to-end using PECO (Planner-Executor Co-Optimization), a reinforcement learning methodology that applies a trajectory-level reward to jointly optimize both roles, fostering seamless collaboration through context-dependent policy updates. Extensive experiments demonstrate that CoDA achieves significant performance improvements over state-of-the-art baselines on complex multi-hop question-answering
benchmarks, and it exhibits strong robustness in long-context scenarios, maintaining stable performance while all other baselines suffer severe degradation, thus further validating the effectiveness of our hierarchical design in mitigating context overload.
Read the full abstract ▾
All authors Xuanzhang Liu, Jianglun Feng, Zhuoran Zhuang, Junzhe Zhao, Maofei Que, Jieting Li, Dianlei Wang, Hao Tong, Ye Chen, Pan Li
Computation and Language · Cross-listed from cs.CV
Xue Li, Xiaonan Song and Henry Hu
Real-world deployment of
Vision-
Language Models (VLMs) is hindered by high computational demands, as existing architectures inefficiently process all
tokens uniformly. We introduce Adaptive
Token Pruning (ATP), a dynamic inference mechanism that retains only the most informative
tokens based on contextual relevance. ATP operates at the
vision-
language interface, assigning a hybrid importance score combining ViT CLS
attention (intra-modal saliency) and CLIP text-image similarity (inter-modal relevance) to keep top-K
tokens for the LLM. Unlike static compression, ATP adapts to each input without modifying the backbone. Proposed as a lightweight gating module, ATP is compatible with popular backbones like BLIP-2, LLaVA, and Flamingo. Preliminary evaluations across VQAv2, GQA, and COCO indicate that ATP reduces inference FLOPs by around 40% and achieves roughly 1.5x speedups in end-to-end latency with negligible accuracy loss (less than 1%). Qualitative analyses suggest ATP preserves visual grounding and enhances interpretability. Beyond efficiency, we investigate robustness under corruptions; observations suggest adaptive
pruning suppresses spurious correlations, improving stability. These findings imply that resource-constrained inference and model reliability are not competing objectives. Finally, we discuss ATP's role in efficient
multimodal edge computing pipelines.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.AI
WebOperator: Action-Aware Tree Search for Autonomous Agents in Web Environment
Mahir Labib Dihan, Tanzima Hashem, Mohammed Eunus Ali et al. (+1 more)
LLM-based
agents often operate in a greedy, step-by-step manner, selecting actions solely based on the current observation without considering long-term consequences or alternative paths. This lack of foresight is particularly problematic in web environments, which are only partially observable-limited to browser-visible content (e.g., DOM and UI elements)-where a single misstep often requires complex and brittle
navigation to undo. Without an explicit backtracking mechanism,
agents struggle to correct errors or systematically explore alternative paths. Tree-search methods provide a principled framework for such structured exploration, but existing approaches lack mechanisms for safe backtracking, making them prone to unintended side effects. They also assume that all actions are reversible, ignoring the presence of irreversible actions-limitations that reduce their effectiveness in realistic web tasks. To address these challenges, we introduce WebOperator, a tree-search framework that enables reliable backtracking and strategic exploration. Our method incorporates a best-first search strategy that ranks actions by both reward estimates and safety considerations, along with a robust backtracking mechanism that verifies the feasibility of previously visited paths before replaying them, preventing unintended side effects. To further guide exploration, WebOperator generates action candidates from multiple, varied
reasoning contexts to ensure diverse and robust exploration, and subsequently curates a high-quality action set by filtering out invalid actions pre-execution and merging semantically equivalent ones. Experimental results on WebArena and WebVoyager demonstrate the effectiveness of WebOperator. On WebArena, WebOperator achieves a state-of-the-art 54.6% success rate with gpt-4o, underscoring the critical advantage of integrating strategic foresight with safe execution.
Read the full abstract ▾
All authors Mahir Labib Dihan, Tanzima Hashem, Mohammed Eunus Ali, Md Rizwan Parvez
Computation and Language · Cross-listed from cs.LG
Reassessing the Role of Supervised Fine-Tuning: An Empirical Study in VLM Reasoning
Yongcan Yu, Lingxiao He, Shuo Lu et al. (+10 more)
Recent advances in
vision-
language models (VLMs)
reasoning have been largely attributed to the rise of reinforcement Learning (RL), which has shifted the community's focus away from the supervised fine-tuning (SFT) paradigm. Many studies suggest that introducing the SFT stage not only fails to improve
reasoning ability but may also negatively impact model training. In this study, we revisit this RL-centric belief through a systematic and controlled comparison of SFT and RL on VLM
Reasoning. Using identical data sources, we find that the relative effectiveness of SFT and RL is conditional and strongly influenced by model capacity, data scale, and data distribution. Contrary to common assumptions, our findings show that SFT plays a crucial role across several scenarios: (1) Effectiveness for weaker models. SFT more reliably elicits
reasoning capabilities in smaller or weaker VLMs. (2) Data efficiency. SFT with only 2K achieves comparable or better
reasoning performance to RL with 20K. (3) Cross-modal transferability. SFT demonstrates stronger generalization across modalities. Moreover, we identify a pervasive issue of deceptive rewards, where higher rewards fail to correlate with better
reasoning accuracy in RL. These results challenge the prevailing "RL over SFT" narrative. They highlight that the role of SFT may have been underestimated and support a more balanced post-training pipeline in which SFT and RL function as complementary components.
Read the full abstract ▾
All authors Yongcan Yu, Lingxiao He, Shuo Lu, Lijun Sheng, Yinuo Xu, Yanbo Wang, Kuangpu Guo, Jianjie Cheng, Meng Wang, Qianlong Xie, Xingxing Wang, Dapeng Hu, Jian Liang
Computation and Language · Cross-listed from cs.LG
How Prompts Move Language Model Behavior: Frames, Salience, and Construal as Semantic Control
Dongseok Kim, Hyoungsun Choi, Mohamed Jismy Aashik Rasool et al. (+1 more)
Prompt engineering is widely used to shape large
language model behavior, yet it is often treated as a practical heuristic rather than as a form of natural-
language control. This paper develops a cognitive-semantic account in which
prompts function as semantic conditions on how a fixed model interprets inputs, foregrounds information, and structures tasks. We formalize this account through three notions – frame activation, salience control, and construal selection – and study them in natural
language inference, claim verification, and multi-hop question answering. Across these settings,
prompts produce measurable changes in label judgments, evidence use, and answer-support organization, showing that
prompt effects differ not only in magnitude but also in semantic direction. The paper therefore reframes prompting as the analysis of how instructions move model behavior, rather than only whether they improve performance.
Read the full abstract ▾
All authors Dongseok Kim, Hyoungsun Choi, Mohamed Jismy Aashik Rasool, Gisung Oh
Computation and Language · Cross-listed from cs.AI
Memoria: A Scalable Agentic Memory Framework for Personalized Conversational AI
Samarth Sarin, Lovepreet Singh, Bhaskarjit Sarmah et al. (+1 more)
Agentic
memory is emerging as a key enabler for large
language models (LLM) to maintain continuity, personalization, and long-term context in extended user interactions, critical capabilities for deploying LLMs as truly interactive and adaptive
agents. Agentic
memory refers to the
memory that provides an LLM with
agent-like persistence: the ability to retain and act upon information across conversations, similar to how a human would. We present Memoria, a modular
memory framework that augments LLM-based conversational systems with persistent, interpretable, and context-rich
memory. Memoria integrates two complementary components: dynamic session-level summarization and a weighted knowledge graph (KG)-based user modelling engine that incrementally captures user traits, preferences, and behavioral patterns as structured entities and relationships. This hybrid architecture enables both short-term dialogue coherence and long-term personalization while operating within the
token constraints of modern LLMs. We demonstrate how Memoria enables scalable, personalized conversational artificial intelligence (AI) by bridging the gap between stateless LLM interfaces and agentic
memory systems, offering a practical solution for industry applications requiring adaptive and evolving user experiences.
Read the full abstract ▾
All authors Samarth Sarin, Lovepreet Singh, Bhaskarjit Sarmah, Dhagash Mehta
Computation and Language
Fine-Tuning Causal LLMs for Text Classification: Embedding-Based vs. Instruction-Based Approaches
Amirhossein Yousefiramandi and Ciaran Cooney
We explore efficient strategies to fine-tune decoder-only Large
Language Models (LLMs) for downstream text classification under resource constraints. Two approaches are investigated: (1) attaching a classification head to a pretrained causal LLM and fine-tuning it on the task, using the LLM's final-
token embedding as a sequence representation, and (2) instruction-tuning the LLM in a
prompt-to-response format for classification. To enable single-GPU fine-tuning of models up to 8B parameters, we combine 4-bit model quantization with Low-Rank Adaptation (LoRA) for parameter-efficient training. Experiments on two patent
benchmarks, a 5-class single-label internal corpus and the public WIPO-Alpha multi-label
dataset with 14 categories, show that the
embedding-head approach matches or exceeds fine-tuned BERT baselines on single-label classification while training 10-30x fewer parameters. Instruction-tuning is competitive only in the multi-label regime, and only with substantially larger trainable budgets of at least 100M parameters. These results demonstrate that directly leveraging the internal representations of causal LLMs, together with efficient fine-tuning techniques, yields strong classification performance under limited computational resources. We discuss the advantages of each approach and outline practical guidelines and future directions for optimizing LLM fine-tuning in classification scenarios.
Read the full abstract ▾
Computation and Language
Modeling Authorial Style in Urdu Novels Using Character Interaction Graphs and Graph Neural Networks
Hassan Mujtaba, Hamza Naveed and Hanzlah Munir
Authorship analysis has traditionally focused on lexical and stylistic cues within text, while higher-level narrative structure remains underexplored, particularly for low-resource
languages such as Urdu. This work proposes a graph-based framework that models Urdu novels as character interaction networks to examine whether authorial style can be inferred from narrative structure alone. Each novel is represented as a graph where nodes correspond to characters and edges denote their co-occurrence within narrative proximity. We systematically compare multiple graph representations, including global structural features, node-level semantic summaries, unsupervised graph
embeddings, and supervised graph neural networks. Experiments on a
dataset of 52 Urdu novels written by seven authors show that learned graph representations substantially outperform hand-crafted and unsupervised baselines, achieving up to 0.857 accuracy under a strict author-aware evaluation protocol.
Read the full abstract ▾
Computation and Language
LexRel: Benchmarking Legal Relation Extraction for Chinese Civil Cases
Yida Cai, Ranjuexiao Hu, Huiyuan Xie et al. (+6 more)
Legal relations serve as an important analytical framework for dispute resolution in civil cases. However, legal relations in Chinese civil cases remain underexplored in the field of legal AI, largely due to the absence of comprehensive schemas. In this work, we first introduce a comprehensive schema for legal relations in civil cases, which contains a hierarchical taxonomy and definitions of arguments. Based on this schema, we formulate a legal relation extraction task and present LexRel, an expert-annotated
benchmark for legal relation extraction in the Chinese civil law domain. We use LexRel to evaluate state-of-the-art large
language models (LLMs) on legal relation extraction, showing that current LLMs exhibit significant limitations in accurately identifying civil legal relations. Furthermore, we demonstrate that explicitly incorporating information about legal relations leads to promising performance gains on other downstream legal AI tasks.
Read the full abstract ▾
All authors Yida Cai, Ranjuexiao Hu, Huiyuan Xie, Chenyang Li, Yun Liu, Yuxiao Ye, Zhenghao Liu, Weixing Shen, Zhiyuan Liu
Computation and Language
Which Pieces Does Unigram Tokenization Really Need?
Sander Land and Yuval Pinter
The Unigram tokenization
algorithm offers a probabilistic alternative to the greedy heuristics of Byte-Pair Encoding. Despite its theoretical elegance, its implementation in practice is complex, limiting its adoption to the SentencePiece package and adapters thereof. We bridge this gap between theory and practice by providing a clear guide to implementation and parameter choices. We also identify a simpler
algorithm that accepts slightly higher training loss in exchange for improved compression.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.CV
Reasoning Within the Mind: Dynamic Multimodal Interleaving in Latent Space
Chengzhi Liu, Yuzhe Yang, Yue Fan et al. (+3 more)
Recent advancements in
Multimodal Large
Language Models (MLLMs) have significantly enhanced cross-modal understanding and
reasoning by incorporating Chain-of-Thought (CoT)
reasoning in the semantic space. Building upon this, recent studies extend the CoT mechanism to the visual modality, enabling models to integrate visual information during
reasoning through external tools or explicit image generation. However, these methods remain dependent on explicit step-by-step
reasoning, unstable perception-
reasoning interaction and notable computational overhead. Inspired by human cognition, we posit that thinking unfolds not linearly but through the dynamic interleaving of
reasoning and perception within the mind. Motivated by this perspective, we propose DMLR, a test-time Dynamic
Multimodal Latent
Reasoning framework that employs confidence-guided latent policy gradient optimization to refine latent think
tokens for in-depth
reasoning. Furthermore, a Dynamic Visual Injection Strategy is introduced, which retrieves the most relevant visual features at each latent think
token and updates the set of best visual patches. The updated patches are then injected into latent think
token to achieve dynamic visual-textual interleaving. Experiments across seven
multimodal reasoning benchmarks and various model architectures demonstrate that DMLR significantly improves
reasoning and perception performance while maintaining high inference efficiency.
Read the full abstract ▾
All authors Chengzhi Liu, Yuzhe Yang, Yue Fan, Qingyue Wei, Sheng Liu, Xin Eric Wang
Computation and Language
Understanding Syllogistic Reasoning in LLMs from Formal and Natural Language Perspectives
Aheli Poddar, Saptarshi Sahoo and Sujata Ghosh
We study syllogistic
reasoning in LLMs from the logical and natural
language perspectives. In process, we explore fundamental
reasoning capabilities of the LLMs and the direction this research is moving forward. To aid in our studies, we use 14 large
language models and investigate their syllogistic
reasoning capabilities in terms of symbolic inferences as well as natural
language understanding. Even though this
reasoning mechanism is not a uniform emergent property across LLMs, the perfect symbolic performances in certain models make us wonder whether LLMs are becoming more and more formal
reasoning mechanisms, rather than making explicit the nuances of human
reasoning.
Read the full abstract ▾
Computation and Language
StruProKGR: A Structural and Probabilistic Framework for Sparse Knowledge Graph Reasoning
Yucan Guo, Saiping Guan, Miao Su et al. (+4 more)
Sparse Knowledge Graphs (KGs) are commonly encountered in real-world applications, where knowledge is often incomplete or limited. Sparse KG
reasoning, the task of inferring missing knowledge over sparse KGs, is inherently challenging due to the scarcity of knowledge and the difficulty of capturing relational patterns in sparse scenarios. Among all sparse KG
reasoning methods, path-based ones have attracted plenty of
attention due to their interpretability. Existing path-based methods typically rely on computationally intensive random walks to collect paths, producing paths of variable quality. Additionally, these methods fail to leverage the structured nature of graphs by treating paths independently. To address these shortcomings, we propose a Structural and Probabilistic framework named StruProKGR, tailored for efficient and interpretable
reasoning on sparse KGs. StruProKGR utilizes a distance-guided path collection mechanism to significantly reduce computational costs while exploring more relevant paths. It further enhances the
reasoning process by incorporating structural information through probabilistic path aggregation, which prioritizes paths that reinforce each other. Extensive experiments on five sparse KG
reasoning benchmarks reveal that StruProKGR surpasses existing path-based methods in both effectiveness and efficiency, providing an effective, efficient, and interpretable solution for sparse KG
reasoning.
Read the full abstract ▾
All authors Yucan Guo, Saiping Guan, Miao Su, Jiyao Wei, Xiaolong Jin, Jiafeng Guo, Xueqi Cheng
Computation and Language
Human-Inspired Learning for Large Language Models via Obvious Record and Maximum-Entropy Method Discovery
Hong Su
Large
Language Models (LLMs) excel at extracting common patterns from large-scale corpora, yet they struggle with rare, low-resource, or previously unseen scenarios-such as niche hardware deployment issues or irregular IoT device behaviors-because such cases are sparsely represented in training data. Moreover, LLMs rely primarily on implicit parametric
memory, which limits their ability to explicitly acquire, recall, and refine methods, causing them to behave predominantly as intuition-driven predictors rather than deliberate, method-oriented learners. Inspired by how humans learn from rare experiences, this paper proposes a human-inspired learning framework that integrates two complementary mechanisms. The first, Obvious Record, explicitly stores cause–result (or question–solution) relationships as symbolic
memory, enabling persistent learning even from single or infrequent encounters. The second, Maximum-
Entropy Method Discovery, prioritizes and preserves methods with high semantic dissimilarity, allowing the system to capture diverse and underrepresented strategies that are typically overlooked by next-
token prediction. Verification on a
benchmark of 60 semantically diverse question–solution pairs demonstrates that the proposed
entropy-guided approach achieves stronger coverage of unseen questions and significantly greater internal diversity than a random baseline, confirming its effectiveness in discovering more generalizable and human-inspired methods.
Read the full abstract ▾
Computation and Language · Cross-listed from cs.AI
AgentSHAP: Interpreting LLM Agent Tool Importance with Monte Carlo Shapley Value Estimation
Miriam Horovicz
LLM
agents that use external tools can solve complex tasks, but understanding which tools actually contributed to a response remains a blind spot. No existing XAI methods address tool-level explanations. We introduce AgentSHAP, the first framework for explaining tool importance in LLM
agents. AgentSHAP is model-agnostic: it treats the
agent as a black box and works with any LLM (GPT, Claude, Llama, etc.) without needing access to internal weights or gradients. Using Monte Carlo Shapley values, AgentSHAP tests how an
agent responds with different tool subsets and computes fair importance scores based on game theory. Our contributions are: (1) the first explainability method for
agent tool attribution, grounded in Shapley values from game theory; (2) Monte Carlo sampling that reduces cost from O(2n) to practical levels; and (3) comprehensive experiments on API-Bank showing that AgentSHAP produces consistent scores across runs, correctly identifies which tools matter, and distinguishes relevant from irrelevant tools. AgentSHAP joins TokenSHAP (for
tokens) and PixelSHAP (for image regions) to complete a family of Shapley-based XAI tools for modern generative AI. Code: https://github.com/GenAISHAP/TokenSHAP.
Read the full abstract ▾
Computation and Language
Coupled Variational Reinforcement Learning for Language Model General Reasoning
Xueru Wen, Jie Lou, Yanjiang Liu et al. (+6 more)
While reinforcement learning has achieved impressive progress in
language model
reasoning, it is constrained by the requirement for verifiable rewards. Recent verifier-free RL methods address this limitation by utilizing the probabilities that LLMs generate reference answers as reward signals. However, these approaches typically sample
reasoning traces conditioned only on the question. This design decouples
reasoning-trace sampling from answer information, leading to inefficient exploration and incoherence between traces and final answers. In this paper, we propose \textitC̱o}upled V̱ariational Ṟeinforcement Ḻearning} (CoVRL), which bridges variational inference and reinforcement learning by coupling prior and posterior distributions through a hybrid sampling strategy. By constructing and optimizing a composite distribution that integrates these two distributions, CoVRL enables efficient exploration while preserving strong thought-answer coherence. Extensive experiments on mathematical and general
reasoning benchmarks show that CoVRL improves performance by 12.4% over the base model and achieves an additional 2.3% improvement over state-of-the-art verifier-free RL baselines, providing a principled framework for enhancing the general
reasoning capabilities of
language models.
Read the full abstract ▾
All authors Xueru Wen, Jie Lou, Yanjiang Liu, Hongyu Lin, Ben He, Xianpei Han, Le Sun, Yaojie Lu, Debing Zhang
Computation and Language
HyperEdit: Unlocking Instruction-based Text Editing in LLMs via Hypernetworks
Yiming Zeng, Jinghan Cao, Zexin Li et al. (+7 more)
Instruction-based text editing is increasingly critical for real-world applications such as code editors (e.g., Cursor), but Large
Language Models (LLMs) continue to struggle with this task. Unlike free-form generation, editing requires faithfully implementing user instructions while preserving unchanged content, as even minor unintended modifications can break functionality. Existing approaches treat editing as generic text generation, leading to two key failures: they struggle to faithfully align edits with diverse user intents, and they often over-edit unchanged regions. We propose HyperEdit to address both issues. First, we introduce hypernetwork-based dynamic adaptation that generates request-specific parameters, enabling the model to tailor its editing strategy to each instruction. Second, we develop difference-aware regularization that focuses supervision on modified spans, preventing over-editing while ensuring precise, minimal changes. HyperEdit achieves a 9%–30% relative improvement in BLEU on modified regions over state-of-the-art baselines, despite utilizing only 3B parameters.
Read the full abstract ▾
All authors Yiming Zeng, Jinghan Cao, Zexin Li, Wanhao Yu, Zhankai Ye, Dawei Xiang, Ting Hua, Xin Liu, Shangqian Gao, Tingting Yu
Computation and Language
NagaNLP: Bootstrapping NLP for Low-Resource Nagamese Creole with Human-in-the-Loop Synthetic Data
Agniva Maiti, Manya Pandey and Murari Mandal
The vast majority of the world's
languages, particularly creoles like Nagamese, remain severely under-resourced in Natural
Language Processing (NLP), creating a significant barrier to their representation in digital technology. This paper introduces NagaNLP, a comprehensive open-source toolkit for Nagamese, bootstrapped through a novel methodology that relies on LLM-driven but human-validated synthetic data generation. We detail a multi-stage pipeline where an expert-guided LLM (Gemini) generates a candidate corpus, which is then refined and annotated by native speakers. This synthetic-hybrid approach yielded a 10K pair conversational
dataset and a high-quality annotated corpus for foundational tasks. To assess the effectiveness of our methodology, we trained both discriminative and generative models. Our fine-tuned XLM-RoBERTa-base model establishes a new
benchmark for Nagamese, achieving a 93.81% accuracy (0.90 F1-Macro) on Part-of-
Speech tagging and a 0.75 F1-Macro on Named Entity Recognition, massively outperforming strong zero-shot baselines. Furthermore, we fine-tuned a Llama-3.2-3B Instruct model, named NagaLLaMA, which demonstrates superior performance on conversational tasks, achieving a Perplexity of 3.85, an order of magnitude improvement over its few-shot counterpart (96.76). We release the NagaNLP toolkit, including all
datasets, models, and code, providing a foundational resource for a previously underserved
language and a reproducible framework for reducing data scarcity in other low-resource contexts.
Read the full abstract ▾