Knowledge catalogue
Gemini 3 Pro was the first model to achieve at least 23% on ARC-AGI-2, which it did in November, 2025 (it actually scored 31%). So the 8-12 month gap between closed and open weights models still seems
arXiv:2511.21945v3 Announce Type: replace Abstract: Generating complete 3D objects under partial occlusions (i.e., amodal scenarios) is a practically important yet challenging problem, as large portio
arXiv:2606.23843v1 Announce Type: new Abstract: Vision-Language Models (VLMs) are typically pre-trained on large-scale image-text datasets to capture semantic correspondences between visual content an
arXiv:2606.24155v1 Announce Type: new Abstract: Existing medical AI benchmarks lack process visibility, atomic skill evaluation, and integrated hallucination detection. We introduce MedBench v5, a red
arXiv:2504.06138v3 Announce Type: replace-cross Abstract: Professional users need tools to help them gain actionable insights from large multimedia collections. Foundation models and AI agents have ra
arXiv:2606.24101v1 Announce Type: cross Abstract: Conventional visual navigation policies often struggle with myopic decision-making and mode collapse in complex environments. While world models offer
🧠 Paradigm II — Agent Foundation Model: world modeling as agent capability. Single-turn, non-agentic environment prediction → tested directly on multi-turn, tool-calling agent tasks. No agentic RL, no
arXiv:2606.24346v1 Announce Type: cross Abstract: Petroleum-engineering search exposes a supervision gap for strong general retrievers: relevant evidence exists in public web text, but domain relevanc
arXiv:2606.23695v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) grounds Large Language Models in external knowledge, yet current evaluations rely on discrete heuristics that suf
arXiv:2606.16845v2 Announce Type: replace-cross Abstract: Small-scale Large Language Models (LLMs) natively default to literal semantic interpretations, making few-shot irony detection a persistent ch
arXiv:2606.24033v1 Announce Type: cross Abstract: Existing low-bit KV-cache quantizers often treat each cached key as a flat vector. Under RoPE, however, a key's contribution to a future attention log
arXiv:2508.10971v2 Announce Type: replace-cross Abstract: Knowledge graphs (KGs) can be enhanced through rule mining; however, the resulting logical rules are often difficult for humans to interpret d
arXiv:2606.24626v1 Announce Type: new Abstract: As autonomous agents tackle increasingly complex multi-step, multi-agent tasks, their execution trajectories have scaled beyond the constraints of even
arXiv:2606.24747v1 Announce Type: new Abstract: Large Language Models (LLMs) achieve strong performance across a growing range of domains, yet their scale poses deployment challenges in applications w
arXiv:2606.24712v1 Announce Type: cross Abstract: Humans effortlessly locate and identify objects by touch alone, even without vision. In contrast, robotic systems rely heavily on vision and struggle
This is the strongest ARC-AGI-2 performance to date by an open-source model. GLM-5.2 from @Zai_org on ARC-AGI (Verified) - ARC-AGI-2: 22.8%, 0.25 - ARC-AGI-1: 77.0%, 0.19 Performance is comparable wit
arXiv:2606.24256v1 Announce Type: new Abstract: Physical interactions follow a long-tailed distribution: a set of common and regular interactions dominates human experience and visual data, while a br
arXiv:2606.23917v1 Announce Type: new Abstract: Advances in generative AI have made image falsification highly realistic, demanding trustworthy authentication systems. Existing forensic detectors can
arXiv:2507.01752v4 Announce Type: replace-cross Abstract: Gradient-based optimization is the workhorse of deep learning, offering efficient and scalable training via backpropagation. However, exposing
arXiv:2606.24094v1 Announce Type: new Abstract: Unifying image clustering across different clustering scenarios remains challenging due to fundamental gaps among tasks. We introduce a Guideline-Driven
arXiv:2606.24477v1 Announce Type: cross Abstract: Video large language models (LLMs) are often constrained by computation and memory budgets, leading them to use reduced frame rates and spatial resolu
arXiv:2606.24066v1 Announce Type: cross Abstract: Speaker recognition has advanced rapidly with large-scale training datasets, yet Vietnamese remains under-resourced, with existing corpora limited in
arXiv:2606.24525v1 Announce Type: new Abstract: GUI agents powered by vision-language models show strong potential for automating digital tasks, yet frequently fail in long-horizon scenarios due to th
arXiv:2606.24370v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly integrated into decision-support roles in business and policy contexts. While prior benchmark studies have
arXiv:2606.21623v1 Announce Type: new Abstract: Instruction-conditioned trajectory prediction is an emerging problem in autonomous driving, where a model predicts the future ego trajectory not only fr
arXiv:2606.20657v1 Announce Type: cross Abstract: Post-training a frontier model is normally weeks of human work: proposing data and recipe changes, launching runs, reading evals, deciding what to kee
arXiv:2606.21884v1 Announce Type: new Abstract: It is tempting to assume any task solvable by a short program can be taught to a model as its chain-of-thought: write the steps out, fine-tune, and the
AI Capex pushback from GS: 'If frontier intelligence can increasingly be developed in the East at a fraction of the cost incurred in the West … then the largest capital allocators are also the ones mo
arXiv:2606.21257v1 Announce Type: new Abstract: OpenPangu models are attractive targets for private and domestic large-language-model deployment, yet their robustness under aggressive post-training qu
arXiv:2606.22278v1 Announce Type: cross Abstract: Ensuring safety of learning-enabled robotic manipulation across diverse embodiments and tasks still requires significant manual engineering. Existing
arXiv:2606.22188v1 Announce Type: new Abstract: Large multi-modal language models are increasingly deployed in high-stakes domains, making well-calibrated uncertainty essential. Traditional Bayesian m
arXiv:2606.22497v1 Announce Type: new Abstract: Microscopic imaging provides essential visual evidence for studying plant biology and pathology at the cellular and subcellular levels. However, existin
arXiv:2606.21398v1 Announce Type: new Abstract: Vision-Language Models (VLMs) for embodied navigation rely on selecting a fixed number of frames from a growing trajectory history. As episodes extend,
The NVIDIA BioNeMo Agent Toolkit enables AI scientists to use structure prediction, molecular generation, docking, sequence analysis, design, and genomics as callable tools , allowing AI agents to exe
CUGA is a lightweight framework from IBM Research for building agentic AI applications, featuring two dozen practical working examples. The framework enables developers to create autonomous AI agents
arXiv:2606.16214v3 Announce Type: replace Abstract: Modern deep learning models remain notoriously prone to overconfidence, limiting their reliability in high-stakes applications. Bayesian methods aim
arXiv:2606.23280v1 Announce Type: new Abstract: Automated Reward Design (ARD) aims to replace manual reward engineering in reinforcement learning with language-driven reward function synthesis. Howeve
arXiv:2606.22836v1 Announce Type: new Abstract: We present Cloak, a training recipe that endows a Vision-Language-Action (VLA) model with zero-shot cross-embodiment transfer by cloaking the end-effect
arXiv:2603.22169v2 Announce Type: replace Abstract: We propose a new Verbal Reinforcement Learning (VRL) framework for interpretable task-level planning in mobile robotic systems operating under execu
arXiv:2606.21885v1 Announce Type: new Abstract: Foundation models have achieved remarkable performance across medical question answering, imaging, and electronic health record (EHR) tasks, yet reliabl
arXiv:2510.15623v2 Announce Type: replace Abstract: Complex query answering (CQA) goes beyond the widely studied link prediction task by addressing more sophisticated queries that require multi-hop re
arXiv:2606.21631v1 Announce Type: cross Abstract: Data curation is a critical part of post-training pipelines for large language models, yet existing tools often treat ingestion, deduplication, synthe
arXiv:2505.09603v2 Announce Type: replace-cross Abstract: Recently, the robotics community has amassed ever larger and more diverse datasets to train generalist policies. However, while these policies
arXiv:2606.22829v1 Announce Type: new Abstract: Intraoperative Adverse Events (IAEs) detection is critical for improving surgical safety, with bleeding being among the most frequent events across many
arXiv:2602.04291v2 Announce Type: replace Abstract: Multi-expert systems, where multiple Large Language Models (LLMs) collaborate to solve complex tasks, are increasingly adopted for high-performance
arXiv:2509.17901v4 Announce Type: replace Abstract: Speech and audio encoders developed over years of community effort are routinely excluded from video understanding pipelines, not because they fail,
arXiv:2606.22950v1 Announce Type: new Abstract: Online learning is crucial for handling complex data streams in big data applications. Recent research has begun to focus on dynamic scenarios, i.e., no
arXiv:2606.23623v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have established a powerful paradigm for generalist robotic manipulation by grounding control into the semantic reas
arXiv:2606.22672v1 Announce Type: new Abstract: Continuous semantic mapping allows autonomous robots to understand both the spatial structure and the semantic content of complex environments. However,
arXiv:2510.26113v2 Announce Type: replace Abstract: Do Video-LLMs have consistent temporal understanding when videos capture the same event from different viewpoints? To study this question, we introd
arXiv:2606.22496v1 Announce Type: cross Abstract: Large language models (LLMs) offer a natural-language interface for interpreting Internet of Things (IoT) sensor data in smart environments; however,
arXiv:2606.22304v1 Announce Type: new Abstract: Recently, several learning paradigms have been introduced to enforce idempotency in generative models. The goal is to ensure that repeated application o
arXiv:2606.22948v1 Announce Type: cross Abstract: As multimodal agents move from interface understanding to real software control, successful trajectory discovery in live desktop environments becomes
arXiv:2606.21197v1 Announce Type: new Abstract: Vision Language Models (VLMs) have demonstrated impressive performance in tasks requiring joint understanding of images and text, such as image captioni
arXiv:2606.15476v2 Announce Type: replace Abstract: Robots operating in homes, warehouses, and other object-rich environments need memory systems that can find specific object instances on demand. Obj
arXiv:2606.21075v1 Announce Type: cross Abstract: Standard Transformers use a single self-attention pathway to model both global dependencies and local patterns, creating tension between long-range st
arXiv:2606.20867v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models enable general-purpose robotic control via large-scale multimodal pretraining, yet their effectiveness under few-sho
arXiv:2606.20661v1 Announce Type: cross Abstract: The integration of external tools has transitioned LLM agents from passive responders to autonomous systems. However, current benchmarks prioritize ex
arXiv:2606.23312v1 Announce Type: new Abstract: Operating in complex real-world environments requires robots to understand their surroundings on a functional semantic level. This demands a detailed mu
arXiv:2601.12879v2 Announce Type: replace Abstract: Extracting sparse circuits from billion-parameter transformers is constrained by O(2^n) search cost and pervasive feature reuse across co-active pat