cs.AI · 2026-10-21 · No. 150
Artificial Intelligence, 2026-10-21.
50 new papers in cs.AI. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
50 entries-
01
Adaptive Multi-Token Prediction: How far should we go?
Emile Anand, Mohammed Rehan Parwani, Xinyuan Cao
cs.AI · cs.LG · eess.SY
Multi-token prediction (MTP) can accelerate language-model decoding by drafting several tokens and verifying them in a single target-model pass. However, the best speculation horizon changes with context: longer drafts offer greater amortization when reliable, but can waste compute on suffixes discarded after a rejection. Unlike autoregressive drafting, where a stopping rule can decide token by token, MTP must fix the horizon before drafting,...
-
02
Semantic Observability for Autonomous Systems Diagnostics
Jitendra Solanki, Anand Todkar, Ziran Min
cs.AI
Modern distributed microservice deployments generate massive volumes of heterogeneous, multi-format observability logs across physically and logically partitioned infrastructure. Isolating root causes during cascading failures remains an arduous task for Site Reliability Engineers (SREs), bottlenecked by manual keyword filtering and disjointed temporal correlation across monitoring dashboards. We introduce Semantic Observability, an...
-
03
Value-Rich, Differently Organized: Measuring Value Expression and Organization in AI-Generated Debates
Maoyuan Li, Haoyuan Li, Zhongsheng Wang, Jiamou Liu
cs.AI
Current evaluation of LLM-generated debate largely focuses on argumentative quality, leaving the value structure of generated arguments underexamined. We introduce a diagnostic metric framework for comparing how values are expressed and organized in human-reference and AI-generated debates. The framework separates value expression, the range of recognizable values a debate invokes, from value organization, how those values are distributed and...
-
04
Evolving Agent Interfaces
Saaket Agashe, Yuzhe Yang, Gonzalo Gonzalez-Pumariega, Bhuvan Chandra Koduru, Chih-Lun Li, Jiachen Yang, Ang Li, Xin...
cs.AI
Computer-use agents are limited by the interfaces humans give them. An agent may type a spreadsheet cell by cell when a single line of code would do, or print all two thousand lines of a file through a shell to find one function. Both the interfaces and the knowledge of when to use each are built by hand, which cannot keep pace with new applications, workflows, and models. We argue that interface design should itself be an agent capability....
-
05
What Do RTL Repair Benchmarks Actually Measure?
Yeyin Eva Zhu, Rishabh Ranawat
cs.AI
RTL repair benchmarks should measure whether a model can fix broken RTL given a specification. To this end, we construct and hash-bind RTLRepairDataset, 5,784 (broken, fixed) training pairs; RTLRepairBench, 459 held-out cases with 374 syntactic and 85 behavioral mutations; and an evaluation harness supporting simulation and formal equivalence. We then use this infrastructure to identify two measurement confounds. First, simulation and formal...
-
06
Beyond the Sequentiality Bottleneck: A Hybrid Diffusion-Autoregressive Framework for Efficient Multimodal Reasoning
Wei Cai, Jian Zhao, Yuchen Yuan, Xuelong Li
cs.AI
Autoregressive models (ARMs) face a "Sequentiality Bottleneck" in reasoning, while diffusion language models (DLMs) offer parallelism but lack semantic coherence. To addresses both limitations, we propose Collaborative Reasoning Asymmetric Learning (CoRAL), a hybrid framework via an Asymmetric Planner-Executor (APE) architecture: a DLM Planner generates K diverse reasoning trajectories in parallel, while a large ARM Executor selects the...
-
07
Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition
Wenjie Lou, Tianshuo Peng, Shiyang Feng, Shengji Tang, Runmin Ma, Ruizhe Chen, Jinxin Shi, Haoyang Peng, Fangchen...
cs.AI
In this report, we introduce Sage-A1.5, a 35B Mixture-of-Experts agentic model that builds on A1 to advance long-horizon problem solving through adaptive capability composition. Compared with A1, Sage-A1.5 learns not only to solve individual tasks, but also to organize work and revise its approach during execution. Sage-A1.5 highlights three improvements: agentic orchestration, verification, and open scientific discovery. First, agentic...
-
08
Evolving Semantic Indexes for Generative Retrieval
Jinming Nian, Jyun-Yu Jiang, Henry Tsai, Yi Fang, Zhirong Liang
cs.AI · cs.IR · cs.LG
Generative retrieval (GR) casts search and recommendation as autoregressive generation of discrete semantic identifiers (SIDs). SID design largely determines GR quality, but its discrete, interdependent components (SID construction, collision handling, decoding, and training) cannot be optimized end-to-end with gradients and remain largely hand-designed. We introduce a framework that improves these components through LLM-driven evolutionary...
-
09
OptScientist: Automated Discovery of Optimization Laws
Ma ZhenYao, Yue Liang, Jian-Qiao Zhu
cs.AI · cs.LG · cs.SC · econ.EM · math.OC
Many scientific phenomena are best understood as outcomes of underlying optimization problems, yet automated scientific discovery has largely focused on recovering equations that describe observed input-output patterns. We introduce OptScientist, a novel automated framework for discovering compact symbolic optimization laws directly from data. OptScientist comprises (i) a model that infers symbolic optimization problems by first learning to...
-
10
Agentic Ensemble with Vision-Language Models for Visual Anomaly Detection
Shangquan Sun, Wenjun Luo, Xu He, Yoonsoo Lee, Shijian Lu
cs.AI
Visual anomaly detection aims to detect abnormal objects, while anomaly samples are often scarce during model training. Assembling multiple pretrained detectors can help by leveraging their transferable image scores and anomaly maps. However, most existing ensemble methods apply fixed fusion rules over saved predictions which struggle when pretrained detectors clearly disagree due to different training objectives, priors, and visual...
-
11
Humans Correct Their Judgments Through Debate, but Weaker AI Models Do Not
Maria Victoria Carro, Denise Alejandra Mester, Facundo Nieto, Oscar Agustin Stanchi, Guido Ernesto Bergman, Giovanni...
cs.AI
How can we effectively oversee AI systems that surpass human intelligence? AI Safety via Debate proposes that two AI systems argue opposing positions and a judge, either a human or a weaker model, decides which is correct. By facilitating supervisors in providing high-quality training signals, debate aims to encourage truthful and safe behavior from AI systems whose capabilities exceed our own. However, human judgment is not neutral: people...
-
12
Beyond 4 bits: Data-Centric Post-training Quantization and Recovery
Rhea Sanjay Sukthanker, Riccardo Grazzi, James Hensman, Pashmina Cameron
cs.AI
Post-training quantization (PTQ) compresses the weights of a trained model with limited compute and data. Down to 4 bits it is near-lossless, but at 2 bits scalar PTQ configurations fall to chance-level accuracy. We study three choices usually treated as implementation details that matter little at 4 bits but decide 2-bit quality: the quantization representation, the calibration data, and the recovery fine-tuning data, across six dense and...
-
13
DecisionPivot: A Diagnostic Benchmark for Decision Models in LLM Systems
Xialie Zhuang, Mingcong Lei, Zunhai Su, Ri Liu, Jianjin Li, Peixian Ma, Jia Pan, Ngai Wong, Guanhua Chen, Chuan-Wei Kuo
cs.AI
Decision models support routing, guardrails, and output verification in LLM systems. However, item accuracy alone cannot establish whether models make correct decisions across controlled variations or assign reliable confidence to their predictions. We introduce DecisionPivot, a diagnostic benchmark comprising 48 tasks and 40,033 items across seven decision roles. Its core track contains 7,800 items from 13 generated task families, organized...
-
14
Proof-Carrying Contracts: Commitments AI Agents Cannot Break
Carlo Ciliberto, Alessandro Rudi, Enrico Tassi
cs.AI · cs.CR · cs.LO · cs.MA · eess.SY
AI agents increasingly act on behalf of people or other agents, purchasing services, exchanging information and delegating work. What an agent must respect is written in instructions, policies or agreements, but text alone cannot make it comply. It may make mistakes, be redirected by what it reads, or pursue different interests. For an economy of such agents to function, parties need agreements they can rely on without inspecting or trusting...
-
15
Exploring Temporal Context-Aware Physical Grounding in Latent World Models
Yilei Man, Yang Chen, Jingcai Guo
cs.AI
Learning latent dynamics is fundamental to building action-conditioned world models for planning. Among existing families, Joint Embedding Predictive Architectures (JEPAs) offer a natural framework by modeling transitions directly in latent space rather than at the pixel level. However, two issues remain in JEPA-based world models: (i) physical grounding lacks explicit aggregation of temporal context and (ii) action planning always overly...
-
16
Claims of emergence in AI should be grounded in information decomposition
Christoph Riedl, Fernando E. Rosas
cs.AI · cs.HC · cs.IT · cs.LG · cs.MA
When a language model develops capabilities not traceable to its individual components, when a multi-agent AI system outperforms each agent on its own, or when a human-AI pair succeeds where either fails in isolation, AI researchers describe it as a case of ``synergy'' or ``emergence''. However, these are usually informal assessments not supported by a theoretical framework, making these claims impossible to compare or even meaningfully...
-
17
Beyond Trajectory Statistics: Community-Consistent Human Mobility Generation via Evolving User-Location Graphs
Bo Liu, Tong Li, Jiapeng Zhang, Zhu Xiao, Zhuo Tang, Kenli Li
cs.AI
Human mobility data are fundamental to Web-based location services and broader urban applications, and mobility generation provides synthetic trajectories when real data are costly or inaccessible. Although recent mobility generators have incorporated collective mobility patterns, they primarily focus on individual trajectories and aggregate population dynamics, leaving evolving community structures formed by user-location interactions...
-
18
Toolflation: Seller Agents Distort Tool Markets Through Spontaneous Generative Engine Optimization
Zhenwei Tang, Di Bai, Jintao Liu, Xidong Wu, Qianfeng Wen, Xinwu Cheng
cs.AI · cs.CL · cs.MA
Agents are starting to sell to other agents. Seller agents write the listings of tools, services and products, and buyer agents read these listings when they choose what to use. We show that seller agents told only to be selected more often practice generative engine optimization on their own, and that this distorts the market for sellers and for buyers. We build tool markets from real data tools on Amazon and Instagram and run every tool on...
-
19
Stratified Sparse Attention: Reducing the Sparse-to-Dense Gap at Extreme Sparsity
Sahil Joshi, Prithvi Dixit, Aryan Katiyar, Agniva Chowdhury, Joseph E. Gonzalez, Anshumali Shrivastava, Aditya...
cs.AI
Large Language Model inference over long contexts is increasingly bottlenecked by the cost of attending to a growing Key-Value (KV) cache. Sparse attention reduces this cost by accessing only a fraction of the cache, but existing methods face a trade-off between deterministic retrieval, which captures high-scoring tokens but discards residual attention mass, and sampling, which improves coverage at the cost of higher variance. Hybrid methods...
-
20
DIVE: Benchmarking Full-Duplex Interaction with Visual Engagement in Continuous Streams
Yuhang Chen, Guangyue Liu, Shilong Liu
cs.AI
Recent advances in multimodal large language models (MLLMs) have substantially improved spoken interaction and video understanding. Existing online and streaming benchmarks primarily assess causal multimodal perception through timestamped question answering, whereas real-world interaction requires models to continuously identify task-relevant multimodal cues and decide whether, when, and how to respond. This leaves open how well current...
-
21
RACE: RoPE-Aligned Attention Compression Based on Query-Key Activation Energies
Irene Tenison, Abdulrahman Alabdulkareem, Daniel Karl I. Weidele, Nima Dehmamy, Dan Gutfreund
cs.AI
Head-dimension compression can reduce the memory required by the key-value (KV) cache, but existing methods often reconstruct full-dimensional representations during inference, adding significant decoding FLOPs overhead. We formalize the design considerations for converting pretrained attention into a natively reduced-dimensional form that reduces both KV-cache storage and attention computation using a calibration set. Specifically, we...
-
22
FALCA: Falsification-Aware Reinforcement Learning with Hierarchical Credit Assignment for Long-Horizon Deep Search Agents
Zijun Dou, Kun Luo, Chaofan Li, Shuqi Lu, Zheng Liu
cs.AI
Long-horizon deep search requires LLM agents to iteratively decompose complex information needs, gather evidence from external sources, and maintain a reliable research state under multiple constraints. A central challenge is that early mistakes can be amplified over the search trajectory: agents may anchor on an incorrect candidate, collect partially supporting evidence, overlook contradictions, and accumulate errors that misguide subsequent...
-
23
S-Agent 2.0: Spatial Reasoning with a General Coding Agent
Rui Gong, Ruoyu Wang, Yalun Dai, Hao Li, Sidi Yang, Qihao Zhao, Hongyu Li, Zeyang Guan, Fangzhou Hong, Zhaoxi Chen,...
cs.AI · cs.CV
Spatial reasoning asks a model to work out where objects are in 3D space, how they relate to one another, and how they move. Agentic VLMs have recently brought large improvements on such tasks by letting the model actively gather evidence with 3D perception tools instead of answering in a single pass. Even so, on many questions the tools return the right evidence and the model still does not arrive at the correct answer: the gap between model...
-
24
Is Sycophancy a Form of In-Context Updating?
Itai Shapira, Myra Cheng, Hadas Orgad
cs.AI
Factual sycophancy, a language model's deference to a user's false beliefs, is usually understood as a bid for the user's approval at the expense of truthfulness. We argue that this framing over-emphasizes the user and propose viewing sycophancy as a failure within the broader and generally useful process of in-context updating, by which models revise their answers based on information supplied during an interaction. We support this view with...
-
25
ChartEvo: Self-Evolving Agent Harnesses for Long-Tail Chart Reasoning
Qi Wang, Peijie Wang, Jun Cao, Haoyang Tong, Xiao-Hui Li, Fei Yin, Cheng-Lin Liu
cs.AI · cs.CV
Multimodal large language models (MLLMs) have substantially advanced chart understanding and reasoning, yet their performance remains highly uneven across chart types. While modern MLLMs perform well on common charts, their accuracy drops sharply on long-tail charts with diverse visual structures and reasoning requirements. Tool-augmented agents can improve complex chart reasoning by introducing specialized visual capabilities, but manually...
-
26
MaD-RSI: Model-as-Data for Recursive Self-Improvement of Agent Harnesses
Yu Cui, Hong He, Ruiqing Yue, Sicheng Pan, Zhuoyu Sun, Yifei Liu, Ting Li, Wenzhuo Zhu, Baohan Huang, Haibin Zhang, Cong Zuo
cs.AI
Recursive self-improvement (RSI) of agent harnesses has emerged as an effective approach to enhancing the capabilities of large language model (LLM) agents. By searching the harness space for a specific LLM on a given dataset, RSI can discover a harness tailored to the capabilities and behavioral characteristics of that model. However, such optimization often creates a strong coupling between the learned harness and its target LLM, limiting...
-
27
Interruption-Aware Recurrent Reinforcement Learning for UAS Navigation Under Temporary Overrides
Christopher Conrad, Deepak Panda, Weisi Guo, Antonios Tsourdos
cs.AI · cs.RO
Safe autonomous navigation of UASs requires tactical deconfliction to address conflicts that strategic coordination cannot prevent. This task is challenging under partial observability, when sensor or traffic data are missing, delayed, or degraded. Geometry-based methods can offer separation assurances but face limitations in dense encounters. Recurrent reinforcement learning has therefore been proposed to learn navigation policies that...
-
28
Robust Representation Learning for Spiking Neural Networks based on Conditional Information Bottleneck
Yuli Zha, Wang Fan, Yixiao Dai, Lingjie Wang, Bang Hu, Wei Cao, Fengzhe Zhang, Fan Zhang
cs.AI
Spiking Neural Networks (SNNs) have shown great potential for low-power neuromorphic computing due to their event-driven processing and spatiotemporal sparsity, but they remain vulnerable to input perturbations, hardware-intrinsic noise, and adversarial attacks in real-world deployment. We attribute this vulnerability partly to input-specific hidden representations learned by standard SNNs trained with conventional cross-entropy loss, as our...
-
29
A Semi-Analytical Inference Framework for Structural Causal Models
Tatiana Boura, Robert Peharz
cs.AI
Structural Causal Models (SCMs) provide a principled framework for observational, interventional, and counterfactual reasoning, all of which rely on probabilistic inference. Existing frameworks either transform SCMs into alternative tractable representations or rely on approximate inference with expressive generative models. We instead perform inference using a message passing scheme directly on the SCM by representing structural mechanisms...
-
30
Polyscene: Scenario Planning as Diversity Planning
Mustafa F. Abdelwahed, Joan Espasa, Felipe Meneguzzi
cs.AI
Scenario planning is a foresight practice in which analysts construct a small set of plausible futures and prepare an organisation for each. Each scenario is a narrative that must be plausible, internally consistent, relevant and challenging, and the set as a whole must be differentiated. The practice remains manual: a workshop (i.e., a facilitated meeting in which a small group of experts writes the narratives) considers only the factors it...
-
31
InsightRAG: Aggregating Distributed Scholarly Judgments for Scientific Survey Generation
Yuquan Lin, Wenshuo Zhang, Yilin Ye, Shuchang Xu, Zhao Chen, Weiqi Wang, Huamin Qu
cs.AI · cs.DL
The rapid growth of scientific literature makes survey writing increasingly challenging, motivating automatic survey generation. Yet, existing systems must search an open space of papers and an open space of concepts with no trace of how the community itself has navigated them, leading to incomplete scope, weak taxonomies, and unreliable citations. Our core insight is that author-written Related Work sections contain the human judgments...
-
32
IECG: Weakly-Supervised Interpretability-Enhancing Foundation Model for Electrocardiogram
Huynh Dang Nguyen, Trong-Thang Pham, Anh-Thuan Do, Phuong-Nhi Dinh, Brijesh Patel, Ngan Le, Van Nguyen
cs.AI
Electrocardiography (ECG) is a cornerstone of cardiac diagnosis, yet most machine learning methods for ECG analysis lack interpretability, limiting their clinical adoption. We present IECG, a weakly-supervised ECG foundation model that jointly performs diagnostic classification and abnormality localization without requiring manual abnormal region annotations, overcoming a long-standing bottleneck in interpretable ECG learning. Central to IECG...
-
33
Slot-JEPA: Object-Structured Prediction Interfaces for Latent World Models
Peisong Wen, Qianqian Xu, Minye Lei, Siran Dai, Yang Liu, Qingming Huang
cs.AI · cs.CV
As a prototypical latent world model, Joint-Embedding Predictive Architectures (JEPAs) provide strong semantic representations. However, dense patch representations expose a redundant interface to downstream predictors, which must implicitly infer spatial organization and temporal correspondence, presenting three challenges. First, object information is distributed across patches, forcing each downstream predictor to redundantly recover...
-
34
TrajEvolve: Trajectory Evolution for Reinforcement Learning with Hindsight Credit Assignment
Yi Wen, Hao Chen, Wanyu Wang, Maolin Wang, Pengyue Jia, Derong Xu, Huize Tan, Yingyi Zhang, Wenlin Zhang, Weihong...
cs.AI
The Reinforcement Learning (RL) paradigm has achieved remarkable success in enhancing the reasoning capabilities of Large Language Models (LLMs). However, when applied to long-horizon tasks, it faces severe challenges: i) Due to the high task complexity, the model struggles to sample enough successful trajectories, resulting in slow convergence. ii) Relying solely on sparse final rewards causes all steps to be updated indiscriminately,...
-
35
Dancing Diffusion: Learning Continuous Programs for Image Generation and Editing
Cui Enhe, Jinyue Yang, Bojian Yin, Guoqi Li
cs.AI · cs.GR
Diffusion models generate samples through a multi-step denoising trajectory, yet condition every step on the same instruction, so the trajectory cannot be addressed at the level of individual operations. We introduce Dancing Diffusion, which assigns each span of timesteps a dedicated continuous condition block and thereby turns conditioning into a latent program. A compiler maps text features into this program, and a fixed pretrained...
-
36
Building Telecom Language Models with RSOD: Domain Data, Specialist Post-Training, Memory-Conscious Distillation, and Networking Evaluation
Anil Srirangapatna Nagesh, Nitin Gupta, Anuj Taneja, Vivek Singh, Shane Eleniak, Theo Zhao, Moulidharan N, Shashank Keshetty
cs.AI · cs.CL · cs.LG
Telecom language models must combine broad networking knowledge with specialist behavior across routing, broadband services, access technologies, and operational diagnostics while remaining practical to deploy as a single model. We present an end-to-end framework that builds source-grounded synthetic training data from structured networking taxonomies and expert graphs, materializes 212,559 routed training records across fourteen networking...
-
37
When Do Humans and Machines Truly Complement Each Other? Formal Conditions and Empirical Evidence for Human-AI Synergy
Andrea Campagner, Federico Cabitza, Vincenzo Crupi
cs.AI
Can humans and artificial intelligence cooperate such that their combination yields greater accuracy than either agent alone? This result, termed \emph{synergy} or \emph{complementarity}, is notoriously difficult to achieve, even when both agents possess significant independent accuracy. In this paper, we present a formal framework to model human-AI interaction through the lens of reliance patterns, which characterize the joint correctness...
-
38
Lost in Recovery: Demystifying Recovery Behavior of Computer-Use Agents via Goal-Oriented Action Replay
Byeonghun Bang, Sangwoo Park, Jinheon Baek, Kangsan Kim, Youngwan Lee, Sung Ju Hwang
cs.AI · cs.CL · cs.LG
Recent advances in large (vision) language models have enabled computer-use agents that automate digital workflows via fine-grained keyboard and mouse operations. However, such agents remain fragile in long-horizon settings, where errors accumulate and executed actions leave unintended side effects, making the ability to recover from failures critical. Yet, existing remedies remain inadequate: environment resets are infeasible in proprietary...
-
39
HypTI: Hyperbolic Geometry-Aware Multimodal Feature Fusion Framework for Dementia Detection Based on Tabular-Image Data
Jianfei Xu, Huizhi Liang, Judith Harrison, John-Paul Taylor, Jaume Bacardit, Varun Ojha
cs.AI
AI-assisted dementia detection usually requires multimodal fusion of diverse data, such as clinical tabular assessments and neuroimaging data. However, most existing approaches perform feature fusion in Euclidean space. This overlooks the geometric structure of representation, including hierarchical organization in clinical assessments and multi-scale anatomical patterns in neuroimaging data. Recent studies have proved hyperbolic space...
-
40
From Comparison to Selection: Enhancing Cross-Scale Generalization of Pointer Networks Trained with Reinforcement Learning for Capacitated Drone Delivery Routing
Sofiene Lassoued, Devara Tertia, Stefan Lier, Andreas Schwung
cs.AI
Unmanned aerial vehicles (UAVs) enable fast, flexible last-mile and emergency delivery, but their routing remains challenging as the number of clients varies. Classical approaches based on mixed-integer programming and heuristics can scale poorly or require problem-specific design, motivating learning-based alternatives. Reinforcement learning is particularly promising, provided a single policy can generalize across problem sizes without...
-
41
Quantum-JEPA: Quantum Born Priors for Latent-Action JEPA World Models
Zhiyuan Liu, Sicheng Liu, Haiyu Wu
cs.AI
A single observation can admit several plausible futures, making uncertainty an essential part of world modelling. However, quantum generative models remain underexplored for this task. We introduce Quantum-JEPA, a latent-action JEPA framework that learns distributions of future representations using a conditional quantum Born prior. Circuit measurements generate binary latent actions, which a shared predictor maps to alternative future...
-
42
ACE-Compression: An Ability-Centric Evaluation of Compressed Large Language Models
Jiahui Ning, Liang Chen, Yuchen Liang, Yi Zhang, Fenglei Fan
cs.AI · cs.CL
Model compression makes large language models deployable, but today's evaluation tells an incomplete story. Standard practice measures retained task performance and stops there. This misses a critical fact: compression perturbs the distributed knowledge embedded across billions of weights, and task scores alone cannot reveal which underlying abilities survive and which quietly collapse. To solve this question, we introduce ACE-Compression, an...
-
43
MMEnterprise-Bench: Evaluating Enterprise Agents over Evolving Multimodal Records
Seohyun Lee, Jongha Kim, Dohwan Ko, Ji Soo Lee, Seoung Choi, Jayun Ku, Hyunwoo J. Kim
cs.AI
Enterprise agents answer questions over workplace records that evolve through team collaboration. As information is shared, corrected, and revised across documents, chats, and figures, multiple versions of the same information accumulate in the workspace. Reliable question answering therefore requires agents to reason over sufficient evidence, resolve cross-source and temporal conflicts, and recognize when the available evidence is...
-
44
Staged Continual Adaptation of Multimodal Foundation Models for Japanese Financial Documents
Genshin Kakimoto, Atsushi Yanagisawa
cs.AI · cs.CL · cs.CV
Staged post-training of multimodal foundation models chains vision-language alignment, reasoning distillation, and domain tuning, yet how each phase trades capabilities against the others remains under-characterised. Tracking an 8.4B-parameter model from an un-fine-tuned baseline through three training phases on Japanese financial documents (evaluated on five text-input benchmarks), we find the phases interact non-additively: math-only...
-
45
Automatic Differentiation with Constraints in Forward and Inverse Problems
Nikolay Yavich, Alexander Ryabov, Viacheslav Naumov, Sayan Ranu, N. M. Anoop Krishnan, Evgeny Burnaev, Vladimir Vanovskiy
cs.AI · math.NA
Approximate solutions of partial differential equations (PDEs) on unstructured grids can be represented on a graph. Numerical optimization involving stiff evolutionary and static PDEs often requires solving a large system of equations, which can be treated as a constraint. In this work, we present a fully differentiable graph-based implicit solver that incorporates this constraint. We evaluate its performance across four key tasks: a) linear...
-
46
Physical Operating-Condition-Guided Vision-Language Matching for Cavitation Diagnosis in Industrial Pipeline Systems
Yu Sha, Junqi Tao, Deyu Wang, Horst Stoecker, Domagoj Vnucec, Andreas Widlf, Kai Zhou
cs.AI
Cavitation diagnosis in industrial pipeline systems is challenging because acoustic responses vary substantially with operating conditions, while similar time-frequency patterns may correspond to different cavitation states. To address this issue, we propose a physical operating-condition-guided vision-language matching (POC-VLM) for cavitation diagnosis. POC-VLM first constructs physical operating-condition-aware semantic descriptions by...
-
47
SAPD: Learning from Agent Experience without Environment Interaction
Juncheng Diao, Haibin Wang, Wei Shi, Yifan Ding, Qingbin Li, Rongxiang Weng, Xunliang Cai
cs.AI · cs.LG
Scaling the training of Large Language Model (LLM) agents typically requires a large collection of executable environments, whose construction, maintenance, and operation constrain the expansion of training experience. Existing offline trajectories offer an opportunity to reuse interaction experience, yet effectively refining the current policy without revisiting these environments remains challenging: supervised fine-tuning (SFT) can lead to...
-
48
COMPASS: Routing via Capability-Oriented Model Profiling with Adaptive Search, Selection, and Tool Use
Amin Rezaei, Pouya Parsa
cs.AI
LLM routing aims to select a capable model at low cost, but specialized routers can require updates as models and tasks change. We propose COMPASS, an agentic router in which an LLM follows a routing recipe, gathers performance and cost evidence through tools, and optionally inspects candidate answers before selecting a model. Our evaluations on LLMRouterBench and DeepSWE show that agentic routing can improve accuracy and lower model-call...
-
49
Privacy-preserving inference on hosted models: a learned input-embedding transform for frozen Qwen3.8-27B
Mahendra Roopa, Srikanth Roopa
cs.AI
Every prompt sent to a hosted language model is readable by whoever runs it. We study whether the prompt can remain on the client while the host receives transformed vectors that stand in for it. A language model converts text to vectors exactly once, with a public lookup table; we perform that step locally and pass the vectors through a small trained network (55.6M parameters) before they leave the perimeter. The host runs a frozen 27B model...
-
50
LatBridge: Efficient and Stable Model Replacement in Latent-Communication Multi-Agent Systems
Jiajun Zhu, Haobo Li, Zixin Chen, Huamin Qu, Wenshuo Zhang
cs.AI
Multi-agent systems rely on communication to complete tasks, and latent communication passes continuous representations to reduce text generation and re-encoding. However, multi-agent systems must reconfigure agent models, whereas existing latent-communication methods are built for fixed model combinations, so their interfaces may fail after replacement. Our pilot study confirms this interface failure and identifies the requirement for...
This edition is part of The Daily Abstract — cs.AI archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.
#D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.