×

Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达,利用大语言模型自动生成交叉领域文献分析简报。

作者: Shao ×
换一批
01.
Nature (Science) 2026-06-10

In situ nanocrystal confinement for efficient blue perovskite LEDs

Metal halide perovskites have emerged as promising semiconductors for light-emitting diodes (LEDs) owing to their excellent luminescence properties1. However, their performance remains limited, primarily owing to the inherent contradiction between ‘high crystallinity’ and ‘small size’ in the in situ synthesis of perovskite nanocrystals on substrates. Here we report efficient blue perovskite LEDs (PeLEDs) achieved via in situ polymerization-driven nanocrystal confinement to synthesize perovskite films composed of high-quality nanocrystals. The in situ-formed polymer network imposes nanoscale spatial constraints during perovskite nanocrystal growth, enabling nanocrystals with small sizes and a high photoluminescence quantum yield of 83%. Furthermore, polymerizable monomers with sufficient coordination sites allow a prolonged lattice rearrangement of perovskite clusters, promoting the crystallinity of the nanocrystals. The synthesized perovskite nanocrystals are utilized in the fabrication of PeLEDs, resulting in an external quantum efficiency of 21.8% at 491 nm, which is among the highest performances in blue PeLEDs. This work simultaneously controls the thermal dynamics of perovskite crystallization and organic ligand reactions, which helps to advance understanding of the effect of ligand engineering on nanocrystal synthesis, benefiting the development of efficient PeLEDs and other optoelectronic technologies. Efficient blue perovskite light-emitting diodes with an external quantum efficiency of 21.8% are achieved through in situ polymerization-driven nanocrystal confinement.

02.
arXiv (CS.CV) 2026-06-11

Task-Aware Structured Memory for Dynamic Multi-modal In-Context Learning

Multi-modal large language models (MLLMs) depend on in-context learning (ICL) for rapid task adaptation, but their scalability is severely limited by finite context windows and the growing cost of key-value (KV) caches in long multi-modal sequences. Existing memory compression approaches typically rely on rigid token removal or sample-dependent importance estimation, which introduces bias, disrupts semantic structure, particularly for visual representations, and yields static memories that cannot adapt to new queries. We introduce TASM (Task-Aware Structured Memory), a training-free framework that addresses these limitations through task-aware, structure-preserving, and dynamically accessible memory construction. TASM employs task-vector guided compression to replace sample-specific signals with a task-level direction that captures shared relevance across demonstrations. To preserve the underlying manifold, it applies semantics-aware token merging via bipartite graph matching, aggregating tokens without destructive pruning. Finally, TASM structures memory into a hierarchy comprising a compact Core Memory and a Latent Bank, facilitating query-adaptive dynamic retrieval. Evaluations confirm TASM maintains high performance under heavy compression, effectively balancing efficiency with adaptability.

03.
arXiv (CS.CL) 2026-06-15

Does the Judge Prefer English? Evaluating Language-Switching Invariance in LLM-as-a-Judge

作者:

Large language models (LLMs) are now widely used as automatic judges for open-ended instruction-following evaluation. This practice is convenient, scalable, and often more semantically aware than reference-based metrics, but it also introduces a new reliability question: does a judge evaluate the quality of an answer, or does it also react to the language in which the comparison is presented? We propose Judge-LS, a lightweight meta-evaluation protocol that transforms LLMBar response-pair items into English, Chinese, and Chinese-English language-switched variants. A reliable judge should preserve its preference under label-preserving language transformations and should not prefer a language when two answers are translation-equivalent. We evaluate four API-accessible judges on the full 419-item LLMBar benchmark, producing 13,408 successful pairwise judgments. Across models, Chinese and language-switched presentations induce 10.7–14.4% preference flips relative to English, and all judges achieve their highest accuracy in English. However, translation-equivalent tie probes do not reveal a systematic English preference: most probes are judged as ties, and non-tie decisions more often favor Chinese. We add confidence intervals, paired significance tests, and an automatic transformation audit with a sensitivity analysis that excludes mechanically flagged high-risk variants. The experiment requires no model training, uses only API calls, and is feasible on modest local hardware.

04.
arXiv (CS.CL) 2026-06-11

ChartFI: Benchmarking Faithfulness and Insightfulness of Chart Descriptions from Multimodal Large Language Models

Chart descriptions are essential for accessibility, cross-modal retrieval, and assisting readers in extracting insights from complex visualizations. As multimodal large language models (MLLMs) are increasingly adopted for automated chart description generation, a critical question arises: how faithfully and insightfully do these models actually describe charts? Current benchmarks fall short on two fronts: existing datasets consist of simple, homogeneous charts paired with shallow, fact-enumerating descriptions; and prevailing metrics fail to capture the multi-faceted nature of description quality. To address these gaps, we present the Chart Faithfulness and Insightfulness Benchmark (ChartFI-Bench). We first summarize four dimensions that characterize high-quality chart descriptions: factual accuracy, salient feature emphasis, domain-informed guidance, and chart-text complementarity. Guided by these dimensions, we construct a high-quality benchmark comprising 896 chart-description pairs, which feature visually complex charts and semantically rich descriptions. Furthermore, we design four aligned evaluation metrics – Faithfulness, Coverage, Informativeness, and Acuity – to systematically assess the quality of descriptions across these dimensions. Experiments conducted on mainstream MLLMs demonstrate the effectiveness of the proposed framework and reveal common weaknesses among existing models.

05.
arXiv (CS.AI) 2026-06-16

Embedded Arena: Iterative Optimization via Hardware Feedback

arXiv:2606.16190v1 Announce Type: cross Abstract: Embedded devices from wildlife monitoring stations to clinical wearables require local AI inference due to latency, communication, or privacy constraints. Optimizing models for heterogeneous microcontrollers (MCUs) requires simultaneously satisfying hard physical constraints on memory, power, and temperature while preserving accuracy, a multidimensional optimization that is today performed manually by experts. We ask whether an LLM agent can autonomously navigate this complex, multi-turn pipeline guided by real hardware feedback, and introduce a hardware-in-the-loop agent arena in which the agent iteratively refines both model and firmware – compiling, flashing, and measuring on real hardware – to enable closed-loop optimization. Frontier models, including Claude Opus 4.7 and Gemini 3.1 Pro, fail entirely without hardware feedback (0% deployment success), whereas our hardware-in-the-loop formulation achieves the first successful deployment within three iterations and can surpass human expert results within seven. This agentic co-optimization achieves 250x compression for vision models with

06.
arXiv (CS.AI) 2026-06-15

VHDLSuite: Unified Pipeline for LLM VHDL Generation with Data Synthesis and Evaluation

arXiv:2606.13735v1 Announce Type: cross Abstract: Large Language Models (LLM) have shown impressive capabilities in Register Transfer Level (RTL) code generation, particularly for Verilog. However, evaluating their performance with other Hardware Description Languages (HDL), especially VHDL, remains limited although its distinct language characteristics, such as stricter semantic rules, introduce evaluation considerations that differ from Verilog. This lack of coverage restricts fully understanding of how well current models generalize across hardware design languages with differing structures and semantics. To address this gap, we introduce VHDLSuite, a benchmark-centered infrastructure for scalable VHDL generation evaluation, integrating automated benchmark synthesis, executable validation, and multi-model diagnostic analysis. First, we propose a data pipeline that automatically converts Verilog designs and their accompanying testbenches into executable VHDL benchmark instances, followed by VUnit/GHDL-based validation to ensure each released task is compilable, runnable, and consistently checkable in the VHDL environment. Second, we introduce VHDLBench, a benchmark with over 200 VHDL problems with complete and validated testbenches across a wide range of complexity levels. Third, we extensively evaluate cutting-edge LLMs and uncover key challenges specific on LLM-aided VHDL generation. Our findings provide important insights and support future work in multi-language hardware design automation.Our data pipeline, benchmark, and evaluation framework will be open-sourced.

07.
arXiv (CS.CL) 2026-06-16

Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale

Efficient and scalable agentic intelligence requires models that can deliver both low-latency responses and strong reasoning capabilities while remaining practical to train, serve, and deploy. In this report, we present Ling-2.6 and Ring-2.6, a family of models designed to address this challenge at scale. Ling-2.6 is optimized for instant response generation and high capability per output token, whereas Ring-2.6 is tailored for deeper reasoning and more advanced agentic workflows. Instead of training from scratch, we upgrade the Ling-2.0 base model through architectural migration pre-training and large-scale post-training. This upgrade is guided by a unified co-design of model architecture, optimization objectives, serving systems, and agent training environments, enabling improvements in both model capability and deployment efficiency. At the architectural level, we introduce a hybrid linear attention design that integrates Lightning Attention with MLA, improving the efficiency of long-context training and decoding. To further enhance token efficiency, we optimize capability per output token through Evolutionary Chain-of-Thought, Linguistic Unit Policy Optimization, bidirectional preference alignment, and shortest-correct-response distillation. For agentic capabilities, we propose KPop, a reinforcement learning framework designed to support stable training of Ring-2.6-1T on large-scale environment-grounded data. KPop improves training efficiency through asynchronous scheduling across coding, search, tool use, and workflow execution, enabling scalable learning from complex agent-environment interactions. Together, Ling-2.6 and Ring-2.6 provide a practical pathway toward efficient, scalable, and open agentic systems. We open-source all checkpoints in the 2.6 family to support further research and development in practical agentic intelligence.

08.
arXiv (CS.LG) 2026-06-15

A Water Efficiency Dataset for African Data Centers

arXiv:2412.03716v3 Announce Type: replace Abstract: Artificial intelligence (AI) computing and data centers consume large amounts of freshwater, both directly for cooling and indirectly for electricity generation. While most attention has been paid to developed countries such as the U.S., this paper presents the first-of-its-kind dataset that combines nation-level weather and electricity generation data to estimate water usage effectiveness for data centers in 41 African countries across five different climate regions. We also use our dataset to evaluate and estimate the water consumption of inference on two large language models (i.e., Llama-3-70B and GPT-4) in 11 selected African countries. Our estimates suggest that writing a 10-page report using Llama-3-70B could consume as much as {0.66 liters} of water, while the water consumption by GPT-4 for the same task may go up to about {59 liters}. For writing a medium-length email of 120-200 words, Llama-3-70B and GPT-4 could consume about {0.13 liters} and {2.9 liters} of water, respectively. All the numbers for generative model inference tasks are based on public information available in 2024, when we initially prepared the analysis. Since then, AI inference systems have improved substantially. For example, recent disclosures suggest that energy efficiency improved by more than 30x between May 2024 and May 2025. Accordingly, our 2024 estimates should be interpreted as historical reference values rather than as representative of current performance. Interestingly, given the same AI model, 9 of the 11 selected African countries consume less water than the global average, mainly because of lower water intensities for electricity generation.

09.
arXiv (CS.CV) 2026-06-16

Learning New Tasks via Reusable Skills: Skill-Compositional Experts for Embodied Continual Learning

Embodied Continual Learning (ECL) aims to enable robots to continually acquire new manipulation tasks while retaining previously learned behaviors under closed-loop control. Compared with conventional continual learning, ECL suffers from more severe catastrophic forgetting. Feature drift accumulated under closed-loop control progressively propagates through sequential decision-making, leading to degradation of previously learned behaviors. A key challenge in ECL lies in structured skill reuse across continually evolving tasks, since existing methods primarily focus on skill learning without explicitly organizing them for coherent task execution. To address this issue, we propose SCE, a Skill-Compositional Experts framework for ECL. SCE builds a skill base via Compositional Skill Grounding (CSG), which decomposes task demonstrations into reusable skills. Based on this, Dual Execution-and-Transition Experts (DETE) enable new task learning through skill composition, where one branch ensures skill execution and the other supports transitions between skills for coherent behavior. Experiments on LIBERO benchmarks and real-world manipulation tasks demonstrate that SCE consistently improves retention and overall task performance. Further feature drift analyses and ablation studies verify the effectiveness of our method. Project website: https://eqcy.github.io/sce/.

10.
medRxiv (Medicine) 2026-06-16

Efficacy of Ergothioneine Supplementation on Postpartum Fatigue, Sleep Quality, and Quality of Life: A Randomized, Double-Blind, Placebo-Controlled Trial

Background: Postpartum asthenia, characterized by severe fatigue, sleep disturbances, and physiological stress, lacks effective targeted interventions. Ergothioneine (EGT) is a unique, naturally occurring antioxidant that actively accumulates in mitochondria, offering a compelling therapeutic rationale for systemic recovery. This study aimed to evaluate the efficacy of EGT in accelerating postpartum functional restoration and alleviating fatigue. Methods: This single-center, randomized, double-blind, placebo-controlled trial enrolled 40 postpartum women (SF-36 total score [≤] 70) who had ceased breastfeeding. Participants were randomized (1:1) to receive either 120 mg/day of EGT or a matched placebo for 30 days. Efficacy was assessed using the SF-36, Pittsburgh Sleep Quality Index (PSQI), Fatigue Scale-14 (FS-14), and Traditional Chinese Medicine (TCM) asthenia scale. To rigorously evaluate the treatment effects, advanced statistical modeling, including Linear Mixed-Effects Models (LMM) and Analysis of Covariance (ANCOVA) adjusted for baseline covariates, was employed. Results: All 40 participants completed the trial. The EGT group demonstrated robust and accelerated functional recovery. Notably, significant improvements in sleep quality (p = 0.0361) and systemic fatigue (p = 0.0059) were observed as early as Day 15. Importantly, EGT yielded a statistically significant between-group superiority in alleviating mental fatigue compared to placebo at Day 15 (p = 0.0313). By Day 30, the EGT cohort exhibited substantial within-group improvements across all primary metrics, including SF-36 (+35.94%, p = 0.0006) and FS-14 (-27.78%, p = 0.0011). Furthermore, as an additional physiological benefit, EGT induced a selective and significant reduction in hepatic transaminases (ALT: -30.42%; AST: -17.44%) within normal limits, a trend not observed in the placebo group. EGT was exceptionally well-tolerated with no treatment-related adverse events. Conclusions: EGT supplementation (120 mg/day) safely accelerates postpartum functional recovery, offering rapid relief from mental fatigue and sleep disturbances within 15 days, while concurrently optimizing hepatic physiological status. These preliminary clinical signals warrant confirmation in larger, adequately powered cohorts. Trial Registration: ChiCTR2500114171; Prospectively registered on 2025-12-08.

11.
arXiv (CS.CV) 2026-06-17

NTIRE 2025 Challenge on Image Super-Resolution (x4): Methods and Results

This paper presents the NTIRE 2025 image super-resolution ($\times$4) challenge, one of the associated competitions of the 10th NTIRE Workshop at CVPR 2025. The challenge aims to recover high-resolution (HR) images from low-resolution (LR) counterparts generated through bicubic downsampling with a $\times$4 scaling factor. The objective is to develop effective network designs or solutions that achieve state-of-the-art SR performance. To reflect the dual objectives of image SR research, the challenge includes two sub-tracks: (1) a restoration track, emphasizes pixel-wise accuracy and ranks submissions based on PSNR; (2) a perceptual track, focuses on visual realism and ranks results by a perceptual score. A total of 286 participants registered for the competition, with 25 teams submitting valid entries. This report summarizes the challenge design, datasets, evaluation protocol, the main results, and methods of each team. The challenge serves as a benchmark to advance the state of the art and foster progress in image SR.

12.
arXiv (CS.CV) 2026-06-19

DiT-JSCC: Rethinking Deep JSCC with Diffusion Transformers and Semantic Representations

Generative joint source-channel coding (GJSCC) has emerged as a new Deep JSCC paradigm for achieving high-fidelity and robust image transmission under extreme wireless channel conditions, such as ultra-low bandwidth and low signal-to-noise ratio. Recent studies commonly adopt diffusion models as generative decoders, but they frequently produce visually realistic results with limited semantic consistency. This limitation stems from a fundamental mismatch between reconstruction-oriented JSCC encoders and generative decoders, as the former lack explicit semantic discriminability and fail to provide reliable conditional cues. In this paper, we propose DiT-JSCC, a novel GJSCC backbone that can jointly learn a semantics-prioritized representation encoder and a diffusion transformer (DiT) based generative decoder, our open-source project aims to promote the future research in GJSCC. Specifically, we design a semantics-detail dual-branch encoder that aligns naturally with a coarse-to-fine conditional DiT decoder, prioritizing semantic consistency under extreme channel conditions. Moreover, a training-free adaptive bandwidth allocation strategy inspired by Kolmogorov complexity is introduced to further improve the transmission efficiency, thereby indeed redefining the notion of information value in the era of generative decoding. Extensive experiments demonstrate that DiT-JSCC consistently outperforms existing JSCC methods in both semantic consistency and visual quality, particularly in extreme regimes.

13.
arXiv (CS.AI) 2026-06-15

Hierarchical ODE: Learning Continuous-Time Physical Prototypes for Early Link Failure Detection

arXiv:2606.14284v1 Announce Type: cross Abstract: Time series prototype learning is fundamentally challenged by observational ambiguity. Discrete architectures fail to resolve this, as they lack the capacity to decouple stochastic noise from continuous dynamics. Furthermore, rigid closed-set assumptions fail to capture unseen diversity. To address these limitations, we propose a hierarchical ordinary differential equation clustering network, which utilizes neural ordinary differential equation to model latent state evolution as a continuous integral curve. This formulation enforces temporal continuity to effectively disentangle smooth feature trends from stochastic noise, while our adaptive hierarchical mechanism autonomously determines the appropriate number of prototypes without rigid prior constraints. Validated on the early link failure detection task with irregularly sampled time series, the proposed method effectively extracts underlying physical prototypes, thereby enabling robust failure detection. Our code is available at https://github.com/NJ-LNN/Hierarchical-ODE.

14.
arXiv (CS.AI) 2026-06-16

CIWI-CKT: Chaos-Informed Wave Interference Feature Fusion and Cross-City Knowledge Transfer for Traffic Flow Forecasting

arXiv:2606.15642v1 Announce Type: cross Abstract: Accurate traffic flow prediction remains challenging in cross-city, data-scarce scenarios where limited historical data hinders model generalisation. The chaotic nature of traffic dynamics, complex spatio-temporal dependencies, and heterogeneous urban networks complicate few-shot learning across cities. Existing deep learning approaches either treat traffic as purely deterministic or lack mechanisms to model wave-like interference patterns essential for cross-regime traffic dynamics. To address these limitations, this paper proposes CIWI-CKT, a novel Chaos-Informed Wave Interference Feature Fusion framework with Cross-City Knowledge Transfer. Our framework introduces three core innovations: chaos-informed wave generation that extracts measurable chaos invariants and models traffic as adaptive wave components; meta-interference processing that captures wave interactions between support and query regimes while producing a predictability score for confidence estimation; and chaos-aware meta-learning that enables efficient cross-city knowledge transfer while preserving chaotic characteristics. We establish theoretical guarantees including chaos-to-wave stability, wave-induced dimension reduction, and meta-learning generalisation bounds. Extensive experiments on four real-world traffic datasets demonstrate that CIWI-CKT significantly outperforms state-of-the-art spatio-temporal graph learning, transfer learning, prompt-based, and few-shot methods, improving prediction accuracy while substantially reducing required training data.

15.
arXiv (CS.CV) 2026-06-11

Reroute, Don't Remove: Recoverable Visual Token Routing for Vision-Language Models

Vision-language models (VLMs) project images into hundreds to thousands of visual tokens, making decoder inference expensive in both attention computation and KV-cache memory. Existing visual-token reduction methods largely follow a rank-and-remove paradigm: they score visual tokens, keep a compact subset, and permanently discard the rest. We show that this irreversible action is fragile because visual-token importance changes across decoder depth; tokens ranked low at one stage may become relevant in later layers, especially for grounding-sensitive queries. We propose Reroute, a training-free plug-in that replaces removal with recoverable routing. At each routing stage, selected vision tokens pass through decoder blocks, while deferred tokens bypass the stage and re-enter the candidate pool at the next routing decision. Reroute reuses existing attention-score ranking rules and stage-wise schedules, preserving the theoretical TFLOPs and KV-cache budget class of the pruning method it augments. Across FastV, PDrop, and Nüwa variants on LLaVA-1.5 and Qwen backbones, reroute improves grounding under aggressive token reduction while maintaining general VQA performance. These results suggest that VLM token reduction should not be viewed only as irreversible pruning, but also as recoverable routing. The code can be found here: https://github.com/elmma/mllm-reroute/

16.
arXiv (CS.CV) 2026-06-16

GraphBEV++: Multi-Modal Feature Alignment for Autonomous Driving

Feature misalignment in BEV perception is a critical yet often overlooked challenge in autonomous driving, especially under calibration uncertainties between LiDAR and camera sensors. To address this issue, we propose a robust multi-modal fusion framework, GraphBEV++, which systematically mitigates projection-induced misalignment. The framework consists of two key modules: LocalAlign-v2 and GlobalAlign-v2. LocalAlign-v2 introduces neighborhood-aware depth features via graph matching to correct local misalignment. It supports both LSS-based and query-based BEV representations, making it compatible with BEVFusion and BEVFormer architectures for consistent cross-paradigm alignment. GlobalAlign-v2 encompasses two variants: Deformable and Diffusion. The Deformable variant addresses global misalignment in LSS-based multi-modal BEV by explicitly learning cross-modal feature offsets. In contrast, the Diffusion variant targets implicit misalignment in query-based BEV by injecting noise to simulate misalignment and employing a denoising process to recover aligned features. Experimental results show that GraphBEV++ achieves state-of-the-art performance under misalignment noise on nuScenes and Waymo subset, improves long-range detection on Argoverse2, and generalizes effectively to the 3D occupancy prediction task, consistently improving occupancy estimation accuracy and robustness under both clean and noisy settings. Furthermore, GraphBEV++ effectively alleviates misalignment issues in end-to-end autonomous driving. Compared with five baselines (UniAD, VAD, FusionAD, MomAD, and WoTE), it demonstrates superior performance in both open-loop (nuScenes) and closed-loop (Bench2Drive and NAVSIM) evaluations across perception, prediction, and planning tasks.

17.
arXiv (CS.CL) 2026-06-11

ASRU: Activation Steering Meets Reinforcement Unlearning for Multimodal Large Language Models

Multimodal large language models (MLLMs) may memorize sensitive cross-modal information during pretraining, making machine unlearning (MU) crucial. Existing methods typically evaluate unlearning effectiveness based on output deviations, while overlooking the generation quality after unlearning. This can easily lead to hallucinated or rigid responses, thereby affecting the usability and safety of the unlearned model. To address this issue, we propose ASRU, a controllable multimodal unlearning framework that incorporates generation quality as a core evaluation objective. ASRU first induces initial refusal behavior through activation redirection, and then optimizes fine-grained refusal boundaries using a customized reward function, thereby achieving a better trade-off between target knowledge unlearning and model utility. Experiments on Qwen3-VL show that ASRU significantly improves unlearning effectiveness (+24.6%) on average and generation quality (5.8X) on average while effectively preserving model utility, using only a small amount of retained supervision data.

18.
arXiv (CS.CV) 2026-06-11

RelayFormer: A Unified Local-Global Attention Framework for Scalable Image and Video Manipulation Localization

Visual manipulation localization (VML) aims to identify tampered regions in images and videos, a task that has become increasingly challenging with the rise of advanced editing tools. Existing methods face two central issues. The first is resolution diversity. Resizing or padding can distort subtle forensic cues and introduce unnecessary computational cost. The second is the difficulty of extending spatial models for images to spatio-temporal inputs in videos, which often results in maintaining separate architectures for the two data types. To address these challenges, we propose RelayFormer, a unified framework that adapts to varying resolutions and naturally handles both static and temporal visual data. RelayFormer partitions inputs into fixed-size sub-images and introduces Global Local Relay (GLR) tokens that propagate structured context through a relay-based attention mechanism. This design enables efficient exchange of global cues, such as semantic or temporal consistency, while preserving fine-grained manipulation artifacts. Unlike prior approaches that depend on uniform resizing or sparse attention, RelayFormer scales to variable resolutions and video sequences with minimal overhead. Experiments across diverse benchmarks demonstrate superior performance and strong efficiency, combining resolution adaptivity without interpolation or excessive padding, unified processing for images and videos, and a favorable balance between accuracy and computational cost. Code is available at~\href{https://github.com/WenOOI/RelayFormer}{https://github.com/WenOOI/RelayFormer}.

19.
arXiv (CS.AI) 2026-06-12

Structured Testbench Generation for LLM-Driven HDL Design and Verification-Oriented Data Curation

arXiv:2606.12983v1 Announce Type: new Abstract: Automated testbench generation has become a critical bottleneck in large language model (LLM)-driven Register Transfer Level (RTL) workflows, where large numbers of candidate designs must be verified rapidly and reliably. Existing prompt-based approaches treat testbench generation as unconstrained code synthesis, yielding stochastic outputs with high token cost, low reproducibility, and insufficient coverage. To address this gap, we present STG, a Structured Testbench Generation framework that exploits the inherent structure of hardware designs to generate deterministic testbenches. As a direct verification tool, STG runs 720x faster than an iterative LLM-based testbench generation flow and higher rate of successful compilation, achieves higher coverage, and reduces false-pass verdicts on incorrect DUTs. STG also helps identify errors in RTL generation benchmarks by exposing faulty benchmark testbenches. As a data curation engine, it is 11x faster than LLM-based filtering on a single CPU core with 127x less energy, and the resulting distilled models provide state-of-the-art performance in our multi-benchmark evaluation. As a test-time scaling oracle, it reduces node count by 14-47\%. Our models are available at https://huggingface.co/collections/AS-SiliconMind/siliconmind-v12.

20.
arXiv (CS.CL) 2026-06-12

Agents' Last Exam

Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional domains. We argue that this gap is largely an evaluation problem: widely used benchmarks lack sustained performance measurement on real and economically valuable workflows. This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long horizon, economically valuable, real world tasks with verifiable outcomes. Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018 (the U.S. federal occupational taxonomy). It is organized around a task taxonomy with 55 sub fields grouped into 13 industry clusters covering 1K+ tasks. Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is below 1%. ALE is designed as a living benchmark: its task pool grows continuously as new workflows and industries are onboarded. More broadly, ALE is intended not merely as another leaderboard, but as an instrument for closing the gap between benchmark success and GDP relevant impact.

21.
arXiv (CS.CV) 2026-06-17

ProCUA-SFT Technical Report

Training computer-use agents (CUAs) – models that interact with graphical desktops through screenshots and keyboard/mouse actions – requires large-scale, diverse trajectory data collected in full desktop environments. The largest public resource, AgentNet (22.5K human trajectories), leads to negative transfer when used for supervised fine-tuning (SFT): continuing training UI-TARS 7B on AgentNet causes OSWorld success rate to fall from 26.3% to 8-10%. We present ProCUA-SFT, a dataset of 3.1M step-level SFT samples distilled from 93K synthetic trajectories across 2,484 application combinations. The dataset is produced by a fully automated pipeline that (i) synthesizes grounded tasks on live desktops seeded with real-world content – 912 spreadsheets from SpreadsheetBench, approximately 10K permissively-licensed presentations from Zenodo10K, and multi-application OSWorld configs – and (ii) verifies each task's feasibility through binary precondition checking before rollout. A single VLM (Kimi-K2.5) serves as goal generator, precondition judge, and trajectory executor, eliminating planner-actor capability gaps. Each trajectory is expanded into step-prefix samples that exactly reproduce the context layout seen at inference time. Fine-tuning UI-TARS 7B on ProCUA-SFT for one epoch yields 45.0% on OSWorld – an 18.7 percentage-point improvement over the base model and over 35% above AgentNet-trained counterparts. A subset of ProCUA was incorporated into the training data for the Nemotron 3 Nano Omni model, contributing to its computer-use capabilities.

22.
arXiv (CS.CL) 2026-06-16

CODA-BENCH: Can Code Agents Handle Data-Intensive Tasks?

Advanced agents are increasingly demonstrating the potential to operate as autonomous engineers, creating a growing demand for evaluation benchmarks that capture the complexity of real-world development. Such environments typically involve both complex code and large-scale data (i.e., file system). However, existing benchmarks usually evaluate code-centric or data-centric capabilities in isolation, leaving a clear gap with real development scenarios. In this paper, we bridge this gap by introducing CODA-BENCH, the first benchmark to jointly evaluate code and data intelligence in a data-intensive environment. We construct a data-intensive Linux sandbox based on the Kaggle ecosystem (containing hundreds of datasets), where agents must actively explore complex file hierarchies to identify relevant resources and generate code for data-driven analytical tasks. CODA-BENCH comprises 1,009 tasks spanning 31 communities, with each task environment containing an average of 980 files, simulating realistic data scale and noise. Evaluations of advanced agents reveal that even top-performing systems struggle to effectively integrate data discovery with code execution, achieving a success rate of only 61.1%. These results highlight a substantial gap in current agentic capabilities for data-intensive tasks and point to promising directions for future research.

23.
arXiv (CS.CV) 2026-06-16

Detect Before You Leap: Mirage Detection in Vision-Language Models

Vision-language models (VLMs) can produce confident visual answers even when the required visual evidence is missing, blank, or unrelated to the question. This failure mode, recently described as mirage (mirage2026), is especially concerning in medical and document VQA, where a plausible but visually ungrounded answer may be mistaken for image-based evidence. We study the complementary problem of pre-release mirage detection: given an image-question pair, determine whether the VLM should answer or abstain before generation. To that end, we propose a novel model-agnostic Text-Conditioned Layer-wise Internal Alignment (TC-LIA) method that probes patch-token representations across the layers of a CLIP ViT-H/14 vision encoder. The key idea is to project layer-wise image patch tokens into the final CLIP embedding space and measure their similarity with the question embedding, thereby tracking whether question-relevant visual evidence emerges across vision layers. TC-LIA summarizes this alignment trajectory using final image-text cosine similarity, late-layer top-k patch-text alignment, early-to-late gain, and layer-wise slope. These features are combined with pixel-statistic based blank/noise detection, zero-shot domain routing, and structured VLM self-assessment in an ensemble. Across five VQA domains with related, unrelated-real, and blank/noise inputs, and across twelve VLM backbones, Qwen2.5-VL-32B achieves the highest three-class detection accuracy of 94.7% with a 3.0% mirage rate, while Qwen2.5-VL-72B achieves 94.6% accuracy with a lower 2.8% mirage rate. Baseline mirage rates span 21.7-66.6%.

24.
arXiv (CS.CV) 2026-06-18

Cosmos 3: Omnimodal World Models for Physical AI

We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-transformers architecture. By supporting highly flexible input-output configurations, Cosmos 3 seamlessly unifies critical modalities for Physical AI – effectively subsuming vision-language models, video generators, world simulators, and world-action models into a single framework. Our evaluation demonstrates that Cosmos 3 establishes a new state-of-the-art across a diverse suite of understanding and generation tasks, demonstrating omnimodal world models as scalable, general-purpose backbones for embodied agents. Our post-trained Cosmos 3 models were ranked as the best open-source Text-to-Image and Image-to-Video models by Artificial Analysis, and the best policy model by RoboArena at the time the technical report was written. To accelerate open research and deployment in Physical AI, we make our code, model checkpoints, curated synthetic datasets, and evaluation benchmark available under the Linux Foundation's OpenMDW-1.1 License at https://github.com/nvidia/cosmos and https://huggingface.co/collections/nvidia/cosmos3. The project website is available at https://research.nvidia.com/labs/cosmos-lab/cosmos3.

25.
arXiv (CS.CL) 2026-06-15

Sentinel: Decoding Context Utilization via Attention Probing for Efficient LLM Context Compression

Retrieval-augmented generation (RAG) often suffers from long and noisy retrieved contexts. Existing context compression methods typically rely on heuristic relevance estimation or supervised compression models rather than on how LLMs utilize retrieved context during inference. We propose Sentinel, a lightweight sentence-level compression framework that decodes inference-time contextual utilization behaviors from head-wise attention patterns of frozen LLMs. To ground supervision in retrieval-dependent answering behavior, Sentinel trains a lightweight probe using QA examples where the model succeeds only when retrieved context is available. Sentinel performs compression using only a single non-autoregressive forward pass without dedicated compression training or autoregressive scoring. Empirically, we find that effective contextual utilization signals remain accessible even in compact proxy models. On LongBench, Sentinel with a 0.5B proxy model achieves up to 5$\times$ compression while attaining question-answering performance competitive with compression methods built on 7B-scale models. Despite being trained only on English QA data, Sentinel also generalizes effectively to Chinese and out-of-domain settings.