Research

• General in-context learning for embodied agents1 paper raises it · 0 address it

  Enables models to adapt to new tasks and environments without retraining, improving robustness in real-world deployment.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Cost and latency in embodied control loops1 paper raises it · 0 address it

  High computational and time costs limit real-time applicability of agentic systems in dynamic environments.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Physical-state verification and failure recovery1 paper raises it · 0 address it

  Agents must reliably detect and correct errors in physical interactions to avoid irreversible failures in real-world tasks.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Long-horizon task preservation in embodied agents1 paper raises it · 0 address it

  Maintaining progress over extended sequences is critical for complex, multi-step tasks in robotics and navigation.

  • Has GPT-6-Astra Achieved Embodied Intelligence?raises
• Open-source stacks for generalist computer-use agents1 paper raises it · 1 addresses it

  Closed systems hinder reproducibility and systematic study of agent design for real-world computer tasks.

  • OpenGCUA: Toward a Fully Open Stack for Generalist Computer Useraises · addresses

Named by an AI from these papers' abstracts.

• Small LLMs10+400.0%
• AI safety36+227.3%
• Training & fine-tuning48+200.0%
• Robotics24+200.0%
• Computer vision37+164.3%
• Image & video generation28+154.5%
• Evals77+148.4%
• RAG10+100.0%

• arXiv cs.AIOct 17

  Are We Evaluating VLM Token Pruning Right? Looking Beyond Accuracy and Token Count

  Geng Li, Guohao Chen, Shilin Shan, Jianfei Yang

  Visual token pruning aims to reduce inference cost while preserving the capabilities of the original VLM. Yet existing evaluations rely on aggregate task accuracy and retained token count as proxies for pruning quality and actual efficiency. We reveal two limitations: 1) Aggregate accuracy can obscure whether a pruning method truly identifies critical visual tokens; many examples remain solvable under arbitrary token removal and models can exploit textual shortcuts when visual evidence is degraded. 2) Retained token count is an unreliable proxy for efficiency, as identical token budgets can incur different end-to-end costs depending on where and how pruning is performed. Consequently, comparisons based on aggregate accuracy and retained token counts fail to recover the global performance-efficiency Pareto frontier across methods. We introduce PAA-Bench, a plug-and-play evaluation framework for existing benchmarks, to evaluate whether pruning succeeds through meaningful token selection and whether this success translates into real speedup. To determine when token choice actually matters, we propose Random-Pass@k to measure how often an example remains correct under random pruning,…

  PDF
• arXiv cs.AIOct 17

  Toward Collective Model Intelligence: Characterizing and Coordinating Models

  SinapisAI Team, Jian-Dong Liu, Qun Dai, Zhendong Du, Guannan Lai +25

  Independently developed AI models accumulate diverse and complementary capabilities, but a growing model collection does not automatically realize Collective AI. Addressing this gap requires understanding model capabilities with limited evaluation, organizing complementary models to collaborate on diverse tasks, and executing model calls efficiently under resource constraints. We present a systematic approach to pursuing Collective Model Intelligence through three connected components. First, Model Eval constructs evidence-based capability profiles by combining query-level performance estimation with cross-domain prediction, reducing the need for exhaustive evaluation. Second, Mio (Models in One) organizes complementary model candidates and supports task-conditioned, cost-aware selection and collaboration as task requirements evolve. Third, Mos (Model Serving OS) separates model selection from execution and provides endpoint scheduling and shared model deployment and distribution. Experiments show lower capability-estimation errors under limited evaluation. Across nine public benchmarks, Mio 1.0 Lite achieves the highest average score and lowest average cost among compared…

  PDF
• arXiv cs.CVOct 17

  Editing Images into Reusable Layers via Compositional Edit-Prior Distillation

  Wen Xue, Yong Liu, Xueheng Li, Si Wu, Yong Xu +3

  High-quality image editing requires precise modification of target regions while preserving the content of unrelated regions. Existing editing models produce flattened outputs that limit precise local manipulation and layer reuse, while decomposition methods typically optimize layer recovery separately from instruction-driven editing. To this end, we introduce LayerEditor, a unified framework that couples instruction-guided image editing with controllable generation of reusable RGBA layers. At its core, In-Context RGB-Alpha Token Modeling jointly represents the edited canvas and an ordered sequence of RGB-Alpha layers within a single token sequence, enabling semantic edits and explicit control over individual layer content. To retain the pretrained model's editing capability during adaptation to layered outputs, we propose Compositional Edit-Prior Distillation, which transfers its editing prior to both the edited canvas and the image composed from the predicted layers. To support training and evaluation, we construct LayerEditor-100K, a large-scale paired image-editing dataset with ordered RGBA layer annotations, and introduce dedicated benchmarks for layered editing. Extensive…

  PDF
• arXiv cs.LGOct 17

  Gaussian Process Bandits with Paired Auxiliary Observations

  Weibo Huang, Qing Feng, Ruihao Zhu, Cheng Hua

  Bayesian optimization (BO) is widely used for expensive black-box optimization, yet observation noise makes each evaluation less informative and slows posterior learning. Standard BO learns from costly outcomes gathered during optimization. In many applications, however, low-cost auxiliary data are available beforehand. We introduce Auxiliary-Informed Gaussian Process Bandit (AI-GPB), a variance-reduction method that combines an offline estimate of the auxiliary mean with a correction coefficient learned from paired online observations. Notably, it exploits correlation between paired observation errors without requiring the auxiliary mean to approximate the objective. We establish finite-sample GP confidence bounds and regret guarantees, including explicit cumulative and simple regret rates for Matérn kernels. Numerical experiments across continuous BO benchmarks with varied landscapes and dimensions show consistent reductions in cumulative and simple regret relative to vanilla GP-UCB, even when within-pair correlation is moderate and the auxiliary mean is estimated from noisy offline data.

  PDF
• arXiv cs.LGOct 17

  How Does Muon Work in Private Deep Learning? Scalable Differentially Private Training via Orthogonalization

  Alexander Gaponov, Grigory Malinovsky, Egor Shulgin, Peter Richtarik

  Training deep neural networks requires optimizers that balance computational efficiency with formal data privacy. While orthogonalized update rules, such as the recently introduced Muon optimizer, have shown strong empirical performance and scalability, their behavior under Differential Privacy (DP) is not yet understood. In this paper, we propose DP-Muon, a differentially private variant of the Muon optimizer. By analyzing Muon's orthogonalized updates through the lens of a Linear Minimization Oracle (LMO), we show that the leading effect of per-example clipping is a positive rescaling of the gradient, to which the LMO is blind, so that only a small directional residual controlled by the centered gradient variance enters the optimization error. This yields an $\mathcal O(K^{-1/4})$ convergence rate, up to logarithmic factors, to a clipping-dependent neighborhood at a fixed noise scale and, with noise calibrated to a fixed $(\varepsilon,δ)$ budget, the standard $(\sqrt d/(n\varepsilon))^{1/2}$ privacy--utility bound of DP-SGD, without assuming bounded gradients. Furthermore, we investigate the hyperparameter landscape of DP-Muon and document how the optimal learning rate,…

  PDF
• arXiv cs.AIOct 17

  Can LLM Agents Resolve Service Failures? Simulating Multi-Turn Complaint Handling under Information Asymmetry

  Wailing Ng, Mengze Hong, Chen Jason Zhang, Shiqi Fan

  LLM agents are widely adopted in service chatbots for rule-guided support and information access, yet emotionally demanding complaint handling still relies heavily on human agents. We introduce ServiceBench, to our knowledge the first benchmark dedicated to complaint handling, with 2,500 scenarios spanning 184 complaint categories and five service industries. We propose Asymmetric Progressive Complaint Simulation (APCS) to simulate realistic complaint handling, together with a justice-based evaluation framework. Our experiments reveal that even capable LLM agents struggle to handle complaints effectively. They lack the ability to proactively gather information and strategically guide conversations toward resolution, often resorting to premature compensation. We further show that targeted post-training with SFT and DPO substantially improves these capabilities, enabling a smaller model to match or surpass much larger models on several outcomes.

  PDF
• arXiv cs.CLOct 17

  GLOSS-MT: Controllable Low-Resource Machine Translation with an Editable Dictionary

  Mich-Seth Owusu, Stephen E. Moore, Akwasi Asare, Lawrence Adu Gyamfi, Paul Azunre +2

  Adapting a pretrained machine translation (MT) model to new terminology usually requires new parallel data and another round of fine-tuning, which is costly for low-resource languages. We ask whether controllability can be added to MT, and the fine-tuning burden reduced, without degrading the pretrained model. GLOSS-MT looks up the source's noun phrases in an offline English-Twi dictionary, replaces every phrase found by its Twi translation in place, and fine-tunes a translation model once to translate these mixed inputs. We apply the same data and pipeline to NLLB-200 (600M, 1.3B, 3.3B), OPUS-MT en-tw and TranslateGemma 4B. On a document-level English-to-Asante Twi benchmark (84 documents, four domains), every fine-tune scores above its pretrained base in chrF2 (+2.85 to +5.48), and the three NLLB fine-tunes are the best open systems (50.12-50.80). A direct controllability test on human-translated Bible passages shows that the three fine-tunes tested write 79-97% of the dictionary terms, and that replacing every dictionary entry by a different term changes the output accordingly (70-97% of the new terms appear) with no retraining. We release the models, data and code.

  PDF
• arXiv cs.LGOct 17

  fxBench: Evaluating and Understanding Formula Suggestions in Spreadsheets

  Sanket Mhatre, Sumit Gulwani, Vu Le, Yasharth Bajpai, Gust Verbruggen

  Spreadsheets are among the most widely used tools for computations, and formulas are their core building blocks. Besides agentic spreadsheet tools, proactive formula suggestions enable significant productivity gains and are now available in popular spreadsheet software (Microsoft Excel and Google Sheets). Such suggestions pose interesting challenges: the right context must be collected and represented for a machine learning model, which must then be able to write a correct formula, in a way that is fast and cheap enough to trigger on every = (or subsequent keystrokes for auto-completion) without taking users out of their editing flow. Surprisingly, there is no public benchmark for formula suggestions on real spreadsheets. In this paper, we therefore (1) introduce fxBench as a manually curated benchmark of 503 formula suggestion tasks based on Sheetpedia, (2) break down the formula suggestion problem into four capabilities (gather context, understand context, understand intent, write formula) that a formula suggestion system requires, and (3) evaluate existing and new approaches to understand bottlenecks in these capabilities that pose interesting research directions.…

  PDF
• arXiv eess.ASOct 17

  ShlokaDhara: A Multi-Speaker Sanskrit Speech Corpus for Learning Meter-Aware Prosody in Text-to-Recitation

  Harshal Kausadikar, Onkar Kishor Susladkar, Prathmesh Santosh Choudhari, Vedangi Deepak Deshpande, Ismini Lorentzu +1

  Sanskrit verse recitation is governed by $\textit{Chandas}$, a strict prosodic framework defined by rhythmic $\textit{Laghu}$ (light) and $\textit{Guru}$ (heavy) syllables. Because existing Sanskrit speech datasets prioritize ASR or conversational TTS, they lack the metrical fidelity required for authentic verse synthesis. We introduce $\textbf{\textit{ShlokaDhara}}$, a 43-hour, multi-speaker Sanskrit speech corpus capturing diverse meters, recitation styles, and melodic variations. We specifically focus on $\textit{Text-to-Recite}$, a synthesis task explicitly requiring both prosodic accuracy and metrical correctness. Benchmarking state-of-the-art TTS architectures via an automated UTMOS (neural MOS) predictor demonstrates that XTTS-v2 achieves the highest perceptual quality. We further validate the XTTS-v2 generated samples using two paradigms: subjective assessments by 16 Sanskrit experts and novel empirical metrics, specifically, syllable-level $\textit{Laghu-Guru}$ duration analysis to objectively quantify structural metrical awareness. Ultimately, this work establishes a robust dataset and evaluation benchmark for prosody-aware computational Sanskrit recitation. Generated…

  PDF
• arXiv cs.AIOct 17

  Where Natural Language Ends in Agents: Executable Skill Composition with MetaSkill

  Meng Shu, Wei He, Hang Zhou, Runke Liu, Kai Han +1

  Agent Skills express open-ended task knowledge in natural language, but composing them requires persistent execution order and upstream-output handoffs. We propose selective formalization at Skill boundaries: coordination relations with stable references and machine-checkable constraints become executable, while substantive task judgments remain within Skills. We introduce MetaSkill, a hybrid representation that preserves complete Skill packages and local instructions while making invocation dependencies and predecessor-text bindings executable through a directed acyclic graph (DAG) runtime. This interface supports explicit coordination without prescribing how each Skill performs its local task. Across 20 multi-Skill artifact tasks and two LLMs, we compare natural-language procedures, prompt-visible graphs, and MetaSkill derived from a common frozen plan per task, using the same designated Skill packages. Additional comparisons evaluate AgentSkillOS with the same available Skill packages as MetaSkill, as well as agent systems using their configured tools. Blinded artifact judgments yield favorable preference point estimates across all reported comparisons, with 95% task-bootstrap…

  PDF
• arXiv cs.CLOct 17

  PV-Decoding: Progressive Verification Decoding with Decode Gating for Diffusion Language Models

  Renxi Liu, Yiming Li, Shengyin Sun, Chen Chen, Xian Wang +6

  Discrete diffusion language models (dLLMs) offer an attractive alternative to autoregressive generation by updating multiple tokens in parallel, but their inference efficiency is still limited by the cost of safely committing tentative predictions. In particular, speculative decoding for dLLMs is challenging because speculative tokens within the same diffusion block may depend on one another, making aggressive acceptance prone to error propagation and strict verification computationally expensive. We present PV-Decoding, a confidence-driven speculative decoding framework with progressive verification for block diffusion inference. PV-Decoding partitions masked positions into directly decoded tokens, speculative tokens, and unresolved tokens using two confidence thresholds, and then verifies speculative tokens with progressively stronger policies, including single-batch, dual-batch, and multi-batch verification. This design exposes a practical accuracy--efficiency trade-off: weaker verification minimizes overhead, whereas stronger verification improves acceptance precision and robustness. Experiments on three diffusion language models across GSM8K, MBPP, and HumanEval show that…

  PDF
• arXiv cs.LGOct 17

  Tensor Network Datamodels for Training Data Attribution

  Farzaneh Heidari, Md Rifat Arefin, Louis Jaburi, Guillaume Rabusseau

  Datamodels predict how a model's behavior changes when different parts of its training data are included or removed. However, linear datamodels assign fixed contributions to training units and cannot represent non-additive joint effects. We introduce Tensor Network Datamodels (TNDMs), which use low-rank tensor networks to model dependencies among training units while keeping the representation compact. TNDMs learn a map from subsets of the training data to the outcome of training on those subsets, allowing them to represent interactions among several training units. We prove that complementary and redundant training groups can induce high-degree subset functions with compact low-rank representations. A single fitted TNDM can then be used for several training data questions, including predicting the effect of removing data, measuring the importance of individual training units, identifying interactions between training units, and selecting useful subsets of the training data. We evaluate TNDMs on controlled tasks designed to contain strong interactions, as well as standard fine-tuning settings for classification and reasoning. On a controlled reasoning task, TNDMs outperform every…

  PDF
• arXiv cs.AIOct 17

  Neuro-Inspired Inhibitory Message Passing for Accurate and Token-Efficient Multi-Agent Reasoning

  Nima H. Siboni, Magdalena Springer, Vahid Rostami

  Multi-agent systems promise to solve problems by bringing together information beyond the reach of any individual agent. Retrieval and tool use can distribute this information across agents, making communication essential to collaboration. A central challenge is keeping crucial evidence visible in each agent's context as messages accumulate. Repeated information can crowd out rare, decisive evidence. Models of cortical circuits suggest a principle for addressing this imbalance: inhibition can selectively suppress shared inputs while preserving sensitivity to weaker signals. We adapt this principle into inhibitory message passing, a training-free communication router that cancels redundant evidence separately for each evidence sentence and receiver. Each round, it forwards evidence sentences restated in agents' messages, delivering each at most once to receivers whose own evidence lacks it and excluding the senders' reasoning and answers. Against standard message passing, accuracy in peer groups rose by 39-77 percentage points across three benchmarks, and tokens per run fell by 10-46%. Gains held across seven open models spanning four to thirty-two billion parameters and two…

  PDF
• arXiv cs.CLOct 17

  DiffuPlan: Editable Planning State for Diffusion Language Model Agents

  Qisheng Zhou, Zhen Xiong, Qiaoyu Tan

  Diffusion language models (dLLMs) are emerging as an alternative to autoregressive (AR) language models, yet dLLM agents largely inherit an interaction protocol designed for AR generation: once intermediate state is written, later observations can only be incorporated by appending more text. This is a poor match to a defining capability of dLLMs -- reconstructing earlier text conditioned on context that appears both before and after it. We ask whether this capability can support a different agent-state paradigm in which intermediate planning state is directly revisable as evidence arrives. We introduce DiffuPlan, a diffusion search agent that maintains a single editable planning state and reconstructs only the affected steps when later evidence resolves an entity or changes the search route. To support variable-length revisions, we introduce Expansion-Partition Marginalization (EPM), which marginalizes over equivalent expansion paths. We further adapt diffusion trajectory RL to multi-turn search. Across four multi-hop QA benchmarks, DiffuPlan achieves 50.24 average F1, outperforming strong AR search agents. More importantly, under identical data and training, editable planning…

  PDF
• arXiv cs.CLOct 17

  When Do Repetition Loops Lock? Truncation Sampling, Bounded Penalties, and a Blind Spot in DRY

  Yichuan Chen, Xin Wang

  Language models sometimes repeat verbatim until the generation limit. We study the exit rather than the entrance: given that a span has already been repeated $n$ times, when is escape still possible? Along the loop the context is deterministic, so loop survival under any sampler can be computed exactly from one forward pass. Pure sampling leaves a positive escape probability at every step; top-$p$ with threshold $π>1/2$ locks a position exactly when the continuation probability reaches $π$. A bounded penalty can leave a position locked; a penalty that outgrows the model's preference for the loop over unseen tokens guarantees escape under top-$p$. The same analysis predicts a blind spot in the widely deployed DRY sampler: its sequence breakers cap the penalty at $λb^{g-1-a}$ for loops whose longest breaker-free run has $g$ tokens. On eleven open models, 66\% of locked units with $g=6$ stay locked under DRY and 1\% with $g=8$. In free generation with llama.cpp, all 109 loops that survive DRY have $g\le 7$. We propose DRY-C, which adds a term that grows exponentially with the number of verbatim recurrences, up to its look-back window. With it, none of 720 structured-output…

  PDF
• arXiv cs.SEOct 17

  Vibenix: Agentic Software Packaging and Maintenance with Nix

  Martim Monis, Martin Schwaighofer, Nuno Saavedra, Alexandra Mendes, Rene Mayrhofer +1

  Implicit dependencies, uncontrolled network access, and other build-environment impurities create intransparent software supply chains. Identifying every dependency affecting a build is a difficult task. Functional package management addresses this through reproducible packaging, ensuring software relies exclusively on declared, hermetically isolated inputs. However, creating build recipes remains complex and time-consuming, posing a significant barrier to adoption for developers and maintainers. We present Vibenix, an open-source AI-powered agent that bridges this gap by automatically generating Nix packaging expressions. Vibenix uses a large language model (LLM) to iteratively refine recipes through a rule-guided build-repair feedback loop. On 466 packaging tasks from our new NixBench dataset, Vibenix produced buildable packages in 95.3% of cases using Claude Haiku 4.5, compared with 5.6% for a state-of-the-art LLM-based single-shot packaging tool. Manual validation of 100 packages revealed that 47% of the buildable packages were initially functionally correct; a post-build refinement process based on the evaluator-optimizer pattern increased functional correctness by 35…

  PDF
• arXiv q-bio.NCOct 17

  Multilingual Brain Decoding Reveals Shared Cortical Semantic Geometry

  Yi Guo, Yong Hu, Michael Kwok-Po Ng, Shuqiang Wang

  Human languages differ markedly in lexical and syntactic form yet can convey similar semantic content. The neural basis of this cross-language semantic commonality remains unclear. Here we developed a multilingual brain decoding framework and analysed brain recordings from 227 participants spanning Chinese, English and French. We found that cortical semantic representations exhibited consistent semantic geometry relationships across languages and individuals, revealing a shared cortical semantic geometry. Importantly, the cross-language consistency of cortical semantic geometry increased from sensorimotor to higher-order association cortex, forming a shared semantic geometry gradient closely aligned with the principal cortical functional gradient. Our work establishes generalizable multilingual brain decoding and identifies stable cortical semantic geometry as a neural basis for semantic commonality across languages.

  PDF
• arXiv cs.AIOct 17

  SQuIREL: A Context-Relative Framework for Measuring Entity Richness in Wikidata

  Michelle Nwachukwu, Albert Meroño-Peñuela, Martim Brandão

  Wikidata is a large, heterogeneous knowledge graph and applying data quality metrics requires context-specific approaches. The "richness" quality metric is currently under-researched and typically restricted to the ontology level, which fails to capture entity-specific structural depth. To address this gap, we introduce SQuIREL, a method that measures richness directly at the entity level using six dimensions: Semantic Specificity, Qualifier Density, Internal Connectivity, Reference Provenance, External Network Connectivity, and Linguistic Accessibility. These dimensions are kept disaggregated to preserve actionability for targeted editorial intervention. We apply this metric to two case studies of oil paintings (Brazilian vs. French) and writers (grouped by citizenship and gender) to demonstrate how it surfaces representational inequalities that may indicate systemic bias. Our metric successfully distinguishes between highly tailored and basic entity profiles, revealing nuances that raw counts and existing aggregate scores obscure. Specifically, it identifies structural imbalances, such as 48% of Brazilian paintings clustering as low-richness stubs against 0% in the French…

  PDF
• arXiv cs.AIOct 17

  Spherical Physics-informed Neural Operator with Multi-scale Coupling for Meteorological Downscaling

  Yiqiang Ye, Yichi Wang, Haodi Zhang, Jiawei Wen, Jiahui Jiang +3

  Meteorological downscaling is crucial for high-resolution regional climate forecasting and disaster early warning. While neural operators have emerged as a promising paradigm for modeling complex spatiotemporal mappings, existing frameworks often struggle with spherical manifold geometric distortions, inherent atmospheric multi-scale coupling mismatches, and lack of explicit atmospheric laws. We propose the Spherical Physics-informed Neural Operator, which utilizes a Spherical Laplacian Decomposition to partition atmospheric fields into hierarchical frequency components, maintaining exact point-wise correspondence across scales. To evaluate these representations at arbitrary locations, we introduce a localized spherical integral operator that approximates continuous kernel transforms via geometry-aware attention. Dynamical consistency is further enforced by embedding differentiable constraints into the learning process. Extensive experiments demonstrate that our framework attains superior accuracy and zero-shot generalization across various meteorological variables and unseen queries, representing a robust and interpretable solution for global-to-regional meteorological…

  PDF
• arXiv cs.CLOct 17

  AtomSD: Procedural Atom-Guided Self-Distillation for LLM Agents

  Bohan Yu, Pengfei Cao, Kai Tian, Chen Han, Shi-Yang Li +6

  Large language models (LLMs) possess broad factual knowledge and strong reasoning capabilities, but solving complex tasks also requires procedural knowledge about how to act. Existing approaches often represent procedural knowledge through trajectories or bundled skills, limiting flexible reuse across interaction histories. This paper introduces Procedural Atoms, fine-grained and independently applicable units that represent Intent, When, Do, and Avoid guidance. Building on this representation, we propose AtomSD, an Atom-guided Self-Distillation framework in which the learner continually extracts Procedural Atoms from its own successful trajectories and distills them into the evolving policy through Atom-conditioned supervision and task-reward optimization. Comprehensive experiments show that AtomSD consistently outperforms strong reward-based and distillation methods, exceeding the state-of-the-art baseline by up to 9.31 points in average performance. Further analyses demonstrate that the proposed Procedural Atoms provide effective guidance, highlighting the value of atom-level procedural knowledge for policy learning.

  PDF