OpenPrint

Time Series Event Detection (TSED) aims to localize semantically meaningful events in time series data, with critical applications in high-stakes domains. Unlike statistical anomalies, events are often defined by natural-language descriptions with internal temporal-logic structures across multiple physical channels. However, in real-world settings, dense event annotations are expensive to obtain, making purely supervised learning difficult. We introduce Language-guided TSED, a setting where a model is given textual event descriptions and must ground them to intervals in multivariate signals with little or no labeled data. To address this problem, we propose Event Logic Tree (ELT), a knowledge representation framework that converts linguistic descriptions into structured temporal logic over signal primitives. Building on ELT, we present SELA, a neuro-symbolic VLM agent framework that iteratively grounds primitives from signal visualizations and composes them under ELT constraints, producing both event intervals and faithful tree-structured explanations. We further release a real-world benchmark across energy and climate domains with expert knowledge and annotations. Experiments show that SELA improves over supervised fine-tuning and existing zero/few-shot time series reasoning baselines.

Self-improving agents are often framed as recursive systems that discover their own improvement procedures. For high-stakes vertical NLP systems, however, the bottleneck is often not autonomy but the placement of expert knowledge: domain specialists can supply rubrics, failure modes, calibration examples, and deployment constraints that an open search loop would otherwise need to rediscover. We present SIRA (self-improving review agent), an expert-bootstrapped agent factory for scientific peer-review support. SIRA keeps the online reviewer and common execution harness fixed, while offline iterations edit only venue-specific artifacts: rubrics, metadata, prompts, templates, calibration rules, benchmark packs, and failure analyses. On a paper-review agent-creation task, SIRA achieves a mean best held-out decision-label accuracy of 0.941 over five runs, compared with 0.865 for a HyperAgents-style open editable-agent baseline under the same dataset split and metric; it also reaches its best candidate in roughly one third as many scored steps. The claim is bounded but sharp: in peer-review support, self-improvement can be strongest when experts shape the search space first and iteration is restricted to auditable, versioned factory artifacts.

As large language models evolve into tool-augmented agents, a central question remains unresolved: when is external tool use actually justified? Existing agent frameworks typically treat tools as ordinary actions and optimize for task success or reward, offering little principled distinction between epistemically necessary interaction and unnecessary delegation. This position paper argues that agents should invoke external tools only when epistemically necessary. Here, epistemic necessity means that a task cannot be completed reliably via the agent's internal reasoning over its current context, without any external interaction. We introduce the Theory of Agent (ToA), a framework that treats agents as making sequential decisions about whether remaining uncertainty should be resolved internally or delegated externally. From this perspective, common agent failure modes (e.g., overthinking and overacting) arise from miscalibrated decisions under uncertainty rather than deficiencies in reasoning or tool execution alone. We further discuss implications for training, evaluation, and agent design, highlighting that unnecessary delegation not only causes inefficiency but can impede the development of internal reasoning capability. Our position provides a normative criterion for tool use that complements existing decision-theoretic models and is essential for building agents that are not only correct, but increasingly intelligent.

With the advancement of AI capabilities, AI reviewers are beginning to be deployed in scientific peer review, yet their capability and credibility remain in question: many scientists simply view them as probabilistic systems without the expertise to evaluate research, while other researchers are more optimistic about their readiness without concrete evidence. Understanding what AI reviewers do well, where they fall short, and what challenges remain is essential. However, existing evaluations of AI reviewers have focused on whether their verdicts match human verdicts (e.g., score alignment, acceptance prediction), which is insufficient to characterize their capabilities and limits. In this paper, we close this gap through a large-scale expert annotation study, in which 45 domain scientists in Physical, Biological, and Health Sciences spent 469 hours rating 2,960 individual criticisms (each targeting one specific aspect of a paper) from human-written and AI-generated reviews of 82 Nature-family papers on correctness, significance, and sufficiency of evidence. On a composite of all three dimensions, a reviewing agent powered by GPT-5.2 scores above each paper's top-rated human reviewer (60.0% vs. 48.2%, p = 0.009), while all three AI reviewers (including Gemini 3.0 Pro and Claude Opus 4.5) exceed the lowest-rated human across every dimension. AI reviewers' accurate criticisms are also more often rated significant and well-evidenced, and surface a distinct 26% of issues no human raises. However, AI reviewers overlap far more than humans do (21% vs. 3% for cross-reviewer pairs), and exhibit 16 recurring weaknesses humans do not share, such as limited subfield knowledge, lack of long context management over multiple files, and overly critical stance on minor issues. Overall, our results position current AI reviewers as complements to, not substitutes for, human reviewers.

We propose a new frontier: Neural Computers (NCs) that unify computation, memory, and I/O of traditional computers in a learned runtime state. Our long-term goal is the Completely Neural Computer (CNC): the mature, general-purpose realization of this emerging machine form, with stable execution, explicit reprogramming, and durable capability reuse. As an initial step, we study whether elementary NC primitives can be learned solely from collected I/O traces, without instrumented program state. Concretely, we instantiate NCs as video models that roll out screen frames from instructions, pixels, and user actions (when available) in CLI and GUI settings. We show that NCs can acquire elementary interface primitives, especially I/O alignment and short-horizon control, while routine reuse, controlled updates, and symbolic stability remain challenging. We outline a roadmap toward CNCs, to establish a new computing paradigm beyond today's agents and conventional computers.

Large language models have achieved remarkable success under the autoregressive paradigm, yet high-quality text generation need not be tied to a fixed left-to-right order. Existing alternatives still struggle to jointly achieve generation efficiency, scalable representation learning, and effective global semantic modeling. We propose Cola DLM, a hierarchical latent diffusion language model that frames text generation through hierarchical information decomposition. Cola DLM first learns a stable text-to-latent mapping with a Text VAE, then models a global semantic prior in continuous latent space with a block-causal DiT, and finally generates text through conditional decoding. From a unified Markov-path perspective, its diffusion process performs latent prior transport rather than token-level observation recovery, thereby separating global semantic organization from local textual realization. This design yields a more flexible non-autoregressive inductive bias, supports semantic compression and prior fitting in continuous space, and naturally extends to other continuous modalities. Through experiments spanning 4 research questions, 8 benchmarks, strictly matched ~2B-parameter autoregressive and LLaDA baselines, and scaling curves up to about 2000 EFLOPs, we identify an effective overall configuration of Cola DLM and verify its strong scaling behavior for text generation. Taken together, the results establish hierarchical continuous latent prior modeling as a principled alternative to strictly token-level language modeling, where generation quality and scaling behavior may better reflect model capability than likelihood, while also suggesting a concrete path toward unified modeling across discrete text and continuous modalities.

Diffusion and flow-based models have become the de facto approaches for generating continuous data, e.g., in domains such as images and videos. Their success has attracted growing interest in applying them to language modeling. Unlike their image-domain counterparts, today's leading diffusion language models (DLMs) primarily operate over discrete tokens. In this paper, we show that continuous DLMs can be made effective with minimal adaptation to the discrete domain. We propose Embedded Language Flows (ELF), a class of diffusion models in continuous embedding space based on continuous-time Flow Matching. Unlike existing DLMs, ELF predominantly stays within the continuous embedding space until the final time step, where it maps to discrete tokens using a shared-weight network. This formulation makes it straightforward to adapt established techniques from image-domain diffusion models, e.g., classifier-free guidance (CFG). Experiments show that ELF substantially outperforms leading discrete and continuous DLMs, achieving better generation quality with fewer sampling steps. These results suggest that ELF offers a promising path toward effective continuous DLMs.

Game world modeling (GWM) and reinforcement learning (RL) are often confounded because research papers rarely quantify how difficult the underlying transition prediction problem is at the declared interface (pixels/tokens/latents with finite history). We propose the Transition Complexity Profile (TCP): a small, reproducible set of metrics that characterizes an environment’s (or gameplay dataset’s) induced transition kernel by (i) intrinsic one-step branching, (ii) interaction-induced uncertainty and opponent influence when observable, and (iii) temporal/spatial dependency span via standardized probe curves. TCP is reported with an explicit reference distribution, protocol stochasticity, and a versioned measurement budget (sampling/resampling and fixed probe compute), enabling comparable numbers across benchmarks. We outline how common game families and modern "neural game engine" domains populate this landscape and call for TCP to become standard benchmark metadata and a required statistic in GWM and RL papers. This paper positions TCP as a critical measurement layer to supplement existing evaluation methods such as downstream return, predictive losses, and qualitative rollouts. By quantifying transition branching, interaction uncertainty, and dependency span at the declared interface, TCP aims to enable interpretable cross-domain and cross-protocol comparisons, facilitate architectural choices, and provide a principled map of environment difficulty for world modeling research.

Synthesizing unstructured research materials into manuscripts is an essential yet under-explored challenge in AI-driven scientific discovery. Existing autonomous writers are rigidly coupled to specific experimental pipelines, and produce superficial literature reviews. We introduce PaperOrchestra, a multi-agent framework for automated AI research paper writing. It flexibly transforms unconstrained pre-writing materials into submission-ready LaTeX manuscripts, including comprehensive literature synthesis and generated visuals, such as plots and conceptual diagrams. To evaluate performance, we present PaperWritingBench, the first standardized benchmark of reverse-engineered raw materials from 200 top-tier AI conference papers, alongside a comprehensive suite of automated evaluators. In side-by-side human evaluations, PaperOrchestra significantly outperforms autonomous baselines, achieving an absolute win rate margin of 50%-68% in literature review quality, and 14%-38% in overall manuscript quality.

Predicting crystal structures from chemical compositions is a fundamental challenge in materials discovery, complicated by complex 3D geometries that distinguish it from fields like protein folding. Here, we present Diffusion-based crystAl Omni (DAO), a pretrain-finetune framework for crystal structure prediction integrating two Siamese foundation models: a structure generator and an energy predictor. The generator is pretrained via a two-stage pipeline on a vast dataset of stable and unstable structures, leveraging the predictor to relax unstable configurations and guide the generative sampling. Across two well-known benchmarks, pretraining significantly enhances performance across multiple backbone architectures. Ablation studies confirm that the synergy between the generator and predictor mutually benefits both components. We further validate DAO on three real-world superconductors (Cr6Os2, Zr16Rh8O4, and Zr16Pd8O4) typically inaccessible to conventional computation. For Cr6Os2, DAO achieves a 100% match rate with experimental references and an atomic-position error of 0.0012 under 20-shot generation, performing over 2000 × faster per iteration than DFT-based structure predictors. These compelling results collectively highlight the potential of our approach for advancing materials science research.

Recursive or looped language models have recently emerged as a new scaling axis by iteratively refining the same model computation over latent states to deepen reasoning. We extend such scaling principle from a single model to multi-agent systems, and ask: Can agent collaboration itself be scaled through recursion? To this end, we introduce RecursiveMAS, a recursive multi-agent framework that casts the entire system as a unified latent-space recursive computation. RecursiveMAS connects heterogeneous agents as a collaboration loop through the lightweight RecursiveLink module, enabling in-distribution latent thoughts generation and cross-agent latent state transfer. To optimize our framework, we develop an inner-outer loop learning algorithm for iterative whole-system co-optimization through shared gradient-based credit assignment across recursion rounds. Theoretical analyses of runtime complexity and learning dynamics establish that RecursiveMAS is more efficient than standard text-based MAS and maintains stable gradients during recursive training. Empirically, we instantiate RecursiveMAS under 4 representative agent collaboration patterns and evaluate across 9 benchmarks spanning mathematics, science, medicine, search, and code generation. In comparison with advanced single/multi-agent and recursive computation baselines, RecursiveMAS consistently delivers an average accuracy improvement of 8.3%, together with 1.2×-2.4× end-to-end inference speedup, and 34.6%-75.6% token usage reduction.

Large language models offer a tempting solution to address the peer review crisis. This position paper argues that today's AI systems should not be used to produce paper reviews. We ground this position in an empirical comparison of human- versus AI-generated ICLR 2026 reviews and an evaluation of the effect of automated paper rewriting on different AI reviewers. We identify two critical issues: 1) AI reviewers exhibit a hivemind effect of excessive agreement within and across papers that reduces perspective diversity. 2) AI review scores are trivially gameable through paper laundering: prompting an LLM to rewrite a paper could significantly increase the scores from AI reviewers, demonstrating that LLM reviewers are easy to game through stylistic changes rather than scientific results. However, non-gameability and review diversity are necessary but not sufficient conditions for automation. We argue that addressing the peer review crisis requires a science of peer review automation -- not general-purpose LLMs deployed without rigorous evaluation.

Development in Artificial Intelligence (AI) has accelerated scientific discovery. Alongside recent AI-oriented Nobel prizes, these trends establish the role of AI tools in science. This advancement raises questions about the potential influences of AI tools on scientists and science as a whole, and highlights a potential conflict between individual and collective benefits. To evaluate, we used a pretrained language model to identify AI-augmented research, with an F1-score of 0.875 in validation against expert-labeled data. Using a dataset of 41.3 million research papers across natural science and covering distinct eras of AI, here we show an accelerated adoption of AI tools among scientists and consistent professional advantages associated with AI usage, but a collective narrowing of scientific focus. Scientists who engage in AI-augmented research publish 3.02 times more papers, receive 4.84 times more citations, and become research project leaders 1.37 years earlier than those who do not. By contrast, AI adoption shrinks the collective volume of scientific topics studied by 4.63% and decreases scientist's engagement with one another by 22.00%. Thereby, AI adoption in science presents a seeming paradox -- an expansion of individual scientists' impact but a contraction in collective science's reach -- as AI-augmented work moves collectively toward areas richest in data. With reduced follow-on engagement, AI tools appear to automate established fields rather than explore new ones, highlighting a tension between personal advancement and collective scientific progress.

If AI displaces human workers faster than the economy can reabsorb them, it risks eroding the very consumer demand firms depend on. We show that knowing this is not enough for firms to stop it. In a competitive task-based model, demand externalities trap rational firms in an automation arms race, displacing workers well beyond what is collectively optimal. The resulting loss harms both workers and firm owners. More competition and "better" AI amplify the excess; wage adjustments and free entry cannot eliminate it. Neither can capital income taxes, worker equity participation, universal basic income, upskilling, or Coasian bargaining. Only a Pigouvian automation tax can. The results suggest that policy should address not only the aftermath of AI labor displacement but also the competitive incentives that drive it.

LLM-driven agents demonstrate strong performance in sequential decision-making but often rely on on-the-fly reasoning, re-deriving solutions even in recurring scenarios. This insufficient experience reuse leads to computational redundancy and execution instability. To bridge this gap, we propose Skill-Pro, a framework that enables agents to autonomously learn reusable procedural skills from interaction experiences without parameter updates. By formalizing a Skill-MDP, Skill-Pro transforms passive episodic narratives into executable Skills defined by activation, execution, and termination conditions to ensure executability. To achieve reliable reusability without capability degradation, we introduce Non-Parametric PPO, which leverages semantic gradients for high-quality candidate generation and a PPO Gate for robust Skill verification. Through score-based maintenance, Skill-Pro sustains compact, high-quality procedural memory. Experimental results across in-domain, cross-task, and cross-agent scenarios demonstrate that Skill-Pro achieves superior reuse rates and significant performance gains with extreme memory compression. Visualized evolutionary trajectories and Skill distributions further reveal how Skill-Pro transparently accumulates, refines, and reuses procedural knowledge to facilitate long-term autonomy.

Large language models are increasingly post-trained with reinforcement learning in verifiable domains such as code and math. Yet, current methods for reinforcement learning with verifiable rewards (RLVR) learn only from a scalar outcome reward per attempt, creating a severe credit-assignment bottleneck. Many verifiable environments actually provide rich textual feedback, such as runtime errors or judge evaluations, that explain why an attempt failed. We formalize this setting as reinforcement learning with rich feedback and introduce Self-Distillation Policy Optimization (SDPO), which converts tokenized feedback into a dense learning signal without any external teacher or explicit reward model. SDPO treats the current model conditioned on feedback as a self-teacher and distills its feedback-informed next-token predictions back into the policy. In this way, SDPO leverages the model's ability to retrospectively identify its own mistakes in-context. Across scientific reasoning, tool use, and competitive programming on LiveCodeBench v6, SDPO improves sample efficiency and final accuracy over strong RLVR baselines. Notably, SDPO also outperforms baselines in standard RLVR environments that only return scalar feedback by using successful rollouts as implicit feedback for failed attempts. Finally, applying SDPO to individual questions at test time accelerates discovery on difficult binary-reward tasks, achieving the same discovery probability as best-of-k sampling or multi-turn conversations with 3x fewer attempts.

This position paper argues that large language model (LLM) reasoning should be studied as latent-state trajectory formation rather than as faithful surface chain-of-thought (CoT). This matters because claims about faithfulness, interpretability, reasoning benchmarks, and inference-time intervention all depend on what the field takes the primary object of reasoning to be. We ask what that object should be once three often-confounded factors are separated and formalize three competing hypotheses: H1, reasoning is primarily mediated by latent-state trajectories; H2, reasoning is primarily mediated by explicit surface CoT; and H0, most apparent reasoning gains are better explained by generic serial compute than by any privileged representational object. Reorganizing recent empirical, mechanistic, and survey work under this framework, and adding compute-audited worked exemplars that factorize surface traces, latent interventions, and matched budget expansions, we find that current evidence most strongly supports H1 as a default working hypothesis rather than as a task-independent verdict. We therefore make two recommendations: the field should treat latent-state dynamics as the default object of study for LLM reasoning, and it should evaluate reasoning with designs that explicitly disentangle surface traces, latent states, and serial compute.

Evaluating AI-generated reviews by verdict agreement is widely recognized as insufficient, yet current alternatives rarely audit which concerns a system identifies, how it prioritizes them, or whether those priorities align with the review rationale that shaped the final assessment. We propose concern alignment, a diagnostic framework that evaluates AI reviews at the concern level rather than only at the verdict level. The framework's core data structure is the match graph, a bipartite alignment between official and AI-generated concerns annotated with match type, severity, and post-rebuttal treatment. From this artifact we derive an evaluation ladder that moves from binary accuracy to concern detection, verdict-stratified behavior, decision-aware calibration, and rebuttal-aware decomposition. In a pilot study of four public AI review systems evaluated in six configurations, concern-level analysis suggests that detection alone does not determine review quality; calibration is often the binding constraint. Systems detect non-trivial fractions of official concerns yet most mark 25--55% of concerns on accepted papers as decisive, where, under our operationalization, no official concern on accepted papers was treated as a decisive blocker. Identical overall verdict accuracy can conceal reject-heavy behavior versus low-recall profiles, and low full-review false decisive rates can partly reflect concern dilution rather than calibrated prioritization. Most systems do not emit a native accept/reject, and inferring it from review tone is method-sensitive, reinforcing the need for concern-level diagnostics that remain stable across inference choices. The contribution is a reusable evaluation framework for auditing which concerns AI reviewers identify, how they weight them, and whether those priorities align with the review rationale that informed the paper's final assessment.

Large language model (LLM) applications such as agents and domain-specific reasoning increasingly rely on context adaptation: modifying inputs with instructions, strategies, or evidence, rather than weight updates. Prior approaches improve usability but often suffer from brevity bias, which drops domain insights for concise summaries, and from context collapse, where iterative rewriting erodes details over time. We introduce ACE (Agentic Context Engineering), a framework that treats contexts as evolving playbooks that accumulate, refine, and organize strategies through a modular process of generation, reflection, and curation. ACE prevents collapse with structured, incremental updates that preserve detailed knowledge and scale with long-context models. Across agent and domain-specific benchmarks, ACE optimizes contexts both offline (e.g., system prompts) and online (e.g., agent memory), consistently outperforming strong baselines: +10.6% on agents and +8.6% on finance, while significantly reducing adaptation latency and rollout cost. Notably, ACE could adapt effectively without labeled supervision and instead by leveraging natural execution feedback. On the AppWorld leaderboard, ACE matches the top-ranked production-level agent on the overall average and surpasses it on the harder test-challenge split, despite using a smaller open-source model. These results show that comprehensive, evolving contexts enable scalable, efficient, and self-improving LLM systems with low overhead.

Large language models (LLMs) are increasingly adapted to downstream tasks via reinforcement learning (RL) methods like Group Relative Policy Optimization (GRPO), which often require thousands of rollouts to learn new tasks. We argue that the interpretable nature of language often provides a much richer learning medium for LLMs, compared to policy gradients derived from sparse, scalar rewards. To test this, we introduce GEPA (Genetic-Pareto), a prompt optimizer that thoroughly incorporates natural language reflection to learn high-level rules from trial and error. Given any AI system containing one or more LLM prompts, GEPA samples trajectories (e.g., reasoning, tool calls, and tool outputs) and reflects on them in natural language to diagnose problems, propose and test prompt updates, and combine complementary lessons from the Pareto frontier of its own attempts. As a result of GEPA's design, it can often turn even just a few rollouts into a large quality gain. Across six tasks, GEPA outperforms GRPO by 6% on average and by up to 20%, while using up to 35x fewer rollouts. GEPA also outperforms the leading prompt optimizer, MIPROv2, by over 10% (e.g., +12% accuracy on AIME-2025), and demonstrates promising results as an inference-time search strategy for code optimization.

Ensuring fairness in machine learning models is critical, particularly in high-stakes domains where biased decisions can lead to serious societal consequences. Existing preprocessing approaches generally lack transparent mechanisms for identifying which features or instances are responsible for unfairness. This obscures the rationale behind data modifications. We introduce FairSHAP, a novel pre-processing framework that leverages Shapley value attribution to improve both individual and group fairness. FairSHAP identifies fairness-critical instances in the training data using an interpretable measure of feature importance, and systematically modifies them through instance-level matching across sensitive groups. This process reduces discriminative risk - an individual fairness metric - while preserving data integrity and model accuracy. We demonstrate that FairSHAP significantly improves demographic parity and equality of opportunity across diverse tabular datasets, achieving fairness gains with minimal data perturbation and, in some cases, improved predictive performance. As a model-agnostic and transparent method, FairSHAP integrates seamlessly into existing machine learning pipelines and provides actionable insights into the sources of bias.Our code is on https://github.com/ZhuMuMu0216/FairSHAP.

Semantic ultra-high-resolution (UHR) image segmentation is essential in remote sensing applications such as aerial mapping and environmental monitoring. Transformer-based models remain challenging in this setting because memory grows quadratically with the number of tokens, limiting either spatial resolution or contextual scope. We introduce CASWiT (Context-Aware Stage-Wise Transformer), a dual-branch Swin-based architecture that injects low-resolution contextual information into fine-grained high-resolution features through lightweight stage-wise cross-attention. To strengthen cross-scale learning, we also propose a SimMIM-style pretraining strategy based on masked reconstruction of the high-resolution image. Extensive experiments on the large-scale FLAIR-HUB aerial dataset demonstrate the effectiveness of CASWiT. Under our RGB-only UHR protocol, CASWiT reaches 66.37% mIoU with a SegFormer decoder, improving over strong RGB baselines while also improving boundary quality. On the URUR benchmark, CASWiT reaches 49.2% mIoU under the official evaluation protocol, and it also transfers effectively to medical UHR segmentation benchmarks. Code and pretrained models are available at https://huggingface.co/collections/heig-vd-geo/caswit.

Self-improving AI systems aim to reduce reliance on human engineering by learning to improve their own learning and problem-solving processes. Existing approaches to self-improvement rely on fixed, handcrafted meta-level mechanisms, fundamentally limiting how fast such systems can improve. The Darwin Gödel Machine (DGM) demonstrates open-ended self-improvement in coding by repeatedly generating and evaluating self-modified variants. Because both evaluation and self-modification are coding tasks, gains in coding ability can translate into gains in self-improvement ability. However, this alignment does not generally hold beyond coding domains. We introduce hyperagents, self-referential agents that integrate a task agent (which solves the target task) and a meta agent (which modifies itself and the task agent) into a single editable program. Crucially, the meta-level modification procedure is itself editable, enabling metacognitive self-modification, improving not only the task-solving behavior, but also the mechanism that generates future improvements. We instantiate this framework by extending DGM to create DGM-Hyperagents (DGM-H), eliminating the assumption of domain-specific alignment between task performance and self-modification skill to potentially support self-accelerating progress on any computable task. Across diverse domains, the DGM-H improves performance over time and outperforms baselines without self-improvement or open-ended exploration, as well as prior self-improving systems. Furthermore, the DGM-H improves the process by which it generates new agents (e.g., persistent memory, performance tracking), and these meta-level improvements transfer across domains and accumulate across runs. DGM-Hyperagents offer a glimpse of open-ended AI systems that do not merely search for better solutions, but continually improve their search for how to improve.

The performance of large language model (LLM) systems depends not only on model weights, but also on their harness: the code that determines what information to store, retrieve, and present to the model. Yet harnesses are still designed largely by hand, and existing text optimizers are poorly matched to this setting because they compress feedback too aggressively. We introduce Meta-Harness, an outer-loop system that searches over harness code for LLM applications. It uses an agentic proposer that accesses the source code, scores, and execution traces of all prior candidates through a filesystem. On online text classification, Meta-Harness improves over a state-of-the-art context management system by 7.7 points while using 4x fewer context tokens. On retrieval-augmented math reasoning, a single discovered harness improves accuracy on 200 IMO-level problems by 4.7 points on average across five held-out models. On agentic coding, discovered harnesses surpass the best hand-engineered baselines on TerminalBench-2. Together, these results show that richer access to prior experience can enable automated harness engineering.

Agent performance increasingly depends on harness engineering, yet harness design is usually buried in controller code and runtime-specific conventions, making it hard to transfer, compare, and study as a scientific object. We ask whether the high-level control logic of an agent harness can instead be externalized as a portable executable artifact. We introduce Natural-Language Agent Harnesses (NLAHs), which express harness behavior in editable natural language, and Intelligent Harness Runtime (IHR), a shared runtime that executes these harnesses through explicit contracts, durable artifacts, and lightweight adapters. Across coding and computer-use benchmarks, we conduct controlled evaluations of operational viability, module ablation, and code-to-text harness migration.

We introduce Prompt Tuned Embedding Classification (PTEC) for classifying companies within an investment firm’s proprietary industry taxonomy, supporting their thematic investment strategy. PTEC assigns companies to the sectors they primarily operate in, conceptualizing this process as a multi-label text classification task. Prompt Tuning, usually deployed as a text-to-text (T2T) classification approach, ensures low computational cost while maintaining high task performance. However, T2T classification has limitations on multi-label tasks due to the generation of non-existing labels, permutation invariance of the label sequence, and a lack of confidence scores. PTEC addresses these limitations by utilizing a classification head in place of the Large Language Models (LLMs) language head. PTEC surpasses both baselines and human performance while lowering computational demands. This indicates the continuing need to adapt state-of-the-art methods to domain-specific tasks, even in the era of LLMs with strong generalization abilities. The proposed method integrates constrained decoding using Trie Search and jointly optimizes a soft prompt along with the classification head, demonstrating improved scalability, efficiency, and classification performance on proprietary and public datasets. Our contributions include adapting Trie Search to prevent repetitive label prediction, introducing PTEC with differential learning rates, and providing empirical evidence that PTEC's performance generalizes well across datasets with varying pretraining knowledge.

Time Series Representation Learning (TSRL) focuses on generating informative representations for various Time Series (TS) modeling tasks. Traditional Self-Supervised Learning (SSL) methods in TSRL fall into four main categories: reconstructive, adversarial, contrastive, and predictive, each with a common challenge of sensitivity to noise and intricate data nuances. Recently, diffusion-based methods have shown advanced generative capabilities. However, they primarily target specific application scenarios like imputation and forecasting, leaving a gap in leveraging diffusion models for generic TSRL. Our work, Time Series Diffusion Embedding (TSDE), bridges this gap as the first diffusion-based SSL TSRL approach. TSDE segments TS data into observed and masked parts using an Imputation-Interpolation-Forecasting (IIF) mask. It applies a trainable embedding function, featuring dual-orthogonal Transformer encoders with a crossover mechanism, to the observed part. We train a reverse diffusion process conditioned on the embeddings, designed to predict noise added to the masked part. Extensive experiments demonstrate TSDE's superiority in imputation, interpolation, forecasting, anomaly detection, classification, and clustering. We also conduct an ablation study, present embedding visualizations, and compare inference speed, further substantiating TSDE's efficiency and validity in learning representations of TS data.

This paper presents CompanyKG (version 2), a large-scale heterogeneous graph developed for fine-grained company similarity quantification and relationship prediction, crucial for applications in the investment industry such as market mapping, competitor analysis, and mergers and acquisitions. CompanyKG comprises 1.17 million companies represented as graph nodes, enriched with company description embeddings, and 51.06 million weighted edges denoting 15 distinct inter-company relations. To facilitate a thorough evaluation of methods for company similarity quantification and relationship prediction, we have created four annotated evaluation tasks: similarity prediction, competitor retrieval, similarity ranking, and edge prediction. We offer extensive benchmarking results for 11 reproducible predictive methods, categorized into three groups: node-only, edge-only, and node-edge. To our knowledge, CompanyKG is the first large-scale heterogeneous graph dataset derived from a real-world investment platform, specifically tailored for quantifying inter-company similarity and relationships.

Multi-view image-to-point label transfer is an effective strategy for 3D semantic segmentation, but its performance largely depends on how predictions from multiple image observations are fused for each 3D point. Most existing pipelines rely on hard voting or handcrafted weighting rules, which do not explicitly learn the reliability of each view under varying geometric and image-quality conditions. In this paper, we introduce DeepChoice, a lightweight view-weighting module for image-guided 3D semantic segmentation. For each visible observation of a 3D point, DeepChoice exploits a compact set of visibility cues, including incidence angle, range, contrast, sharpness, signal-to-noise ratio, and saturation, to predict normalized per-view weights used to aggregate 2D semantic class probabilities into final 3D point-wise predictions. The method is sensor-agnostic, requires no meshing, and can be integrated as a drop-in replacement for standard multi-view fusion rules. Experiments on the full GridNet-HD benchmark show that DeepChoice improves over hard voting by 3.85 mIoU points and over mean-probability fusion by 1.26 mIoU points, while reducing the gap with the AnyView oracle upper bound. The largest gains are observed on thin and difficult classes such as conductors, pylons, and insulators. Furthermore, a complementary evaluation on the Images&PointClouds Cultural Heritage dataset shows that the proposed weighting strategy remains beneficial under a very different acquisition context and scene structure, yielding a 1.55 mIoU point improvement over hard voting and a 0.48 mIoU point improvement over mean-probability fusion. A compact Transformer variant provides the best trade-off between accuracy and model size, outperforming a larger MLP-based alternative. These results show that learning how to weight views is a simple yet effective way to strengthen image-guided 3D semantic segmentation pipelines. Code is publicly available at https://huggingface.co/heig-vd-geo/DeepChoice.

Recent generative and tool-using AI systems can surface a large volume of candidates at low marginal cost, yet only a small fraction can be checked carefully. This creates a decoder-side bottleneck: downstream decision-makers must form reliable posteriors from many public records under scarce attention. We formalize this regime via Attention-Constrained Inference (ACI), in which a cheap screening stage processes $K$ records and an expensive verification stage can follow up on at most $B$ of them. Under Bayes log-loss, we study the maximum achievable reduction in posterior uncertainty per window, which we call \emph{epistemic throughput}. Our main result is a ``JaKoB'' scaling law showing that epistemic throughput has a baseline term that grows linearly with verification and prevalence, and an additional \emph{information-leverage} term that scales as $\sqrt{JKB}$, where $J$ summarizes screening quality. Thus, expanding cheap screening can nonlinearly amplify scarce verification, even when informative records are rare. We further show that this scaling is tight in a weak-screening limit, and that in the sparse-verification regime ($B \ll K$), substantial leverage requires heavy-tailed score distributions; for light-tailed scores the amplification is only logarithmic.

As AI systems evolve from isolated predictors into complex, heterogeneous ecosystems of foundation models and specialized adapters, distinguishing genuine behavioral novelty from functional redundancy becomes a critical governance challenge. Here, we introduce a statistical framework for auditing model uniqueness based on In-Silico Quasi-Experimental Design (ISQED). By enforcing matched interventions across models, we isolate intrinsic model identity and quantify uniqueness as the Peer-InExpressible Residual (PIER), i.e. the component of a target’s behavior strictly irreducible to any stochastic convex combination of its peers, with vanishing PIER characterizing when such a routing-based substitution becomes possible. We establish the theoretical foundations of ecosystem auditing through three key contributions. First, we prove a fundamental limitation of observational logs: uniqueness is mathematically non-identifiable without intervention control. Second, we derive a scaling law for active auditing, showing that our adaptive query protocol achieves minimax-optimal sample efficiency (dσ²γ⁻²log(Nd/δ)). Third, we demonstrate that cooperative game-theoretic methods, such as Shapley values, fundamentally fail to detect redundancy. We implement this framework via the DISCO (Design-Integrated Synthetic Control) estimator and deploy it across diverse ecosystems, including computer vision models (ResNet/ConvNeXt/ViT), large language models (BERT/RoBERTa), and city-scale traffic forecasters. These results move trustworthy AI beyond explaining single models: they establish a principled, intervention-based science of auditing and governing heterogeneous model ecosystems.

3D object detection is a crucial research topic in computer vision, which usually uses 3D point clouds as input in conventional setups. Recently, there is a trend of leveraging multiple sources of input data, such as complementing the 3D point cloud with 2D images that often have richer color and fewer noises. However, due to the heterogeneous geometrics of the 2D and 3D representations, it prevents us from applying off-the-shelf neural networks to achieve multimodal fusion. To that end, we propose Bridged Transformer (BrT), an end-to-end architecture for 3D object detection. BrT is simple and effective, which learns to identify 3D and 2D object bounding boxes from both points and image patches. A key element of BrT lies in the utilization of object queries for bridging 3D and 2D spaces, which unifies different sources of data representations in Transformer. We adopt a form of feature aggregation realized by point-to-patch projections which further strengthen the interaction between images and points. Moreover, BrT works seamlessly for fusing the point cloud with multi-view images. We experimentally show that BrT surpasses state-of-the-art methods on SUN RGB-D and ScanNetV2 datasets.