Inferring physical actions from visual observations is a fundamental capability for advancing machine intelligence in the physical world. Achieving this requires large-scale, open-vocabulary video action datasets that span broad domains. We introduce Action100M, a large-scale dataset constructed from 1.2M Internet instructional videos (14.6 years of duration), yielding O(100 million) temporally localized segments with open-vocabulary action supervision and rich captions. Action100M is generated by a fully automated pipeline that (i) performs hierarchical temporal segmentation using V-JEPA 2 embeddings, (ii) produces multi-level frame and segment captions organized as a Tree-of-Captions, and (iii) aggregates evidence with a reasoning model (GPT-OSS-120B) under a multi-round Self-Refine procedure to output structured annotations (brief/detailed action, actor, brief/detailed caption). Training VL-JEPA on Action100M demonstrates consistent data-scaling improvements and strong zero-shot performance across diverse action recognition benchmarks, establishing Action100M as a new foundation for scalable research in video understanding and world modeling.

We envision a proactive multi-modal assistant system which gives users real-time step-by-step guidance on a procedural task, autonomously deciding when to interrupt, and how to coach. However, progress is limited by the absence of large-scale, cross-domain benchmarks that reflect realistic conditions, particularly the common case in which users deviate from the expected step sequence. We address this gap with four contributions: (1) we release EgoProactive, a large-scale wearable-egocentric dataset for proactive procedural assistance with explicit Out-of-Plan (OOP) annotations and recovery steps; (2) we augment five established benchmarks (Ego4D, EPIC-KITCHENS, EgoExo4D, HoloAssist, HowTo100M) into Pro2Bench under a unified proactive-guidance schema; (3) we propose a decoupled planner–interaction architecture specialized for procedural state, visual cues, and recovery injection; (4) we introduce a post-training recipe that transfers across model families, validated by cross-backbone replication on Llama 4 and Qwen-3.6-VL. In extensive experiments, our trained Llama-4 system substantially improves objective intervention quality over strong proprietary baselines (Claude Opus 4.6, Gemini 3.1 Pro, GPT 5.2) and open-weight baselines (Qwen3 VL 235B) baselines across all six datasets. Oracle-plan experiments further show that, when plan quality is controlled, the trained duplex model produces high-quality guidance and large gains on Out-of-Plan recovery.

We introduce VL-JEPA, a vision-language model built on a Joint Embedding Predictive Architecture (JEPA). Instead of autoregressively generating tokens as in classical VLMs, VL-JEPA predicts continuous embeddings of the target texts. By learning in an abstract representation space, the model can focus on task-relevant semantics while abstracting away surface-level linguistic variability. In a strictly controlled comparison against standard token-space VLM training with the same vision encoder and training data, VL-JEPA achieves stronger performance while having 50% fewer trainable parameters. At inference time, a lightweight text decoder is invoked only when needed to translate VL-JEPA predicted embeddings into text. We show that VL-JEPA natively supports selective decoding that can reduce the number of decoding operations by approximately 2.85× while maintaining similar performance compared to dense non-adaptive uniform decoding. Beyond generation, the embedding-space formulation naturally supports open-vocabulary classification, text-to-video retrieval, and discriminative VQA without any architecture modification. On eight video classification and eight video retrieval datasets, the average performance of VL-JEPA surpasses that of CLIP, SigLIP2, and Perception Encoder. At the same time, the model achieves comparable performance to classical VLMs (InstructBLIP, QwenVL) on four VQA datasets: GQA, TallyQA, POPE, and POPEv2, despite only having 1.6B parameters.

Agents must infer action outcomes and select actions that maximize a reward signal indicating how close the goal is to being reached. Supervised learning of reward models could introduce biases inherent to training data, limiting generalization to novel goals and environments. In this paper, we investigate whether well-defined world state representations alone can enable accurate reward prediction across domains. To address this, we introduce StateFactory, a factorized representation method that transforms unstructured observations into a hierarchical object-attribute structure using language models. This structured representation allows rewards to be estimated naturally as the semantic similarity between the current state and the goal state under hierarchical constraint. Overall, the compact representation structure induced by StateFactory enables strong reward generalization capabilities. We evaluate on RewardPrediction, a new benchmark dataset spanning five diverse domains and comprising 2,454 unique action-observation trajectories with step-wise ground-truth rewards. Our method shows promising zero-shot results against both VLWM-critic and LLM-as-a-Judge reward models, achieving 60

Large language models (LLMs) trained on massive multilingual datasets hint at the formation of interlingual constructs–a shared subspace in the representation space. However, evidence regarding this phenomenon is mixed, leaving it unclear whether these models truly develop unified interlingual representations, or present a partially aligned constructs. We explore 31 diverse languages varying on their resource-levels, typologies, and geographical regions; and find that multilingual LLMs exhibit inconsistent cross-lingual alignments. To address this, we propose an interlingual representation framework identifying both the shared interlingual semantic subspace and fragmented components, existed due to representational limitations. We introduce Interlingual Local Overlap (ILO) score to quantify interlingual alignment by comparing the local neighborhood structures of high-dimensional representations. We utilize ILO to investigate the impact of single-language fine-tuning on the interlingual representations in multilingual LLMs. Our results indicate that training exclusively on a single language disrupts the alignment in early layers, while freezing these layers preserves the alignment of interlingual representations, leading to improved cross-lingual generalization. These results validate our framework and metric for evaluating interlingual representation, and further underscore that interlingual alignment is crucial for scalable multilingual learning.

Humans are known to have an internal "world model" that enables us to carry out action planning based on world states. AI agents need to have such a world model for action planning as well. It is not clear how current AI models, especially generative models, are able to learn such world models and carry out procedural planning in diverse environments. We introduce WorldPrediction, a video-based benchmark for evaluating world modeling and procedural planning capabilities of different AI models. In contrast to prior benchmarks that focus primarily on low-level world modeling and robotic motion planning, WorldPrediction is the first benchmark that emphasizes actions with temporal and semantic abstraction. Given initial and final world states, the task is to distinguish the proper action (WorldPrediction-WM) or the properly ordered sequence of actions (WorldPrediction-PP) from a set of counterfactual distractors. This discriminative task setup enable us to evaluate different types of world models and planners and realize a thorough comparison across different hypothesis. The benchmark represents states and actions using visual observations. In order to prevent models from exploiting low-level continuity cues in background scenes, we provide "action equivalents" - identical actions observed in different contexts - as candidates for selection. This benchmark is grounded in a formal framework of partially observable semi-MDP, ensuring better reliability and robustness of the evaluation. We conduct extensive human filtering and validation on our benchmark and show that current frontier models barely achieve 57 whereas humans are able to solve both tasks perfectly.

The first International AI Safety Report comprehensively synthesizes the current evidence on the capabilities, risks, and safety of advanced AI systems. The report was mandated by the nations attending the AI Safety Summit in Bletchley, UK. Thirty nations, the UN, the OECD, and the EU each nominated a representative to the report's Expert Advisory Panel. A total of 100 AI experts contributed, representing diverse perspectives and disciplines. Led by the report's Chair, these independent experts collectively had full discretion over the report's content.

The widespread application of LLMs across various tasks and fields has necessitated the alignment of these models with human values and preferences. Given various approaches of human value alignment, there is an urgent need to understand the scope and nature of human values injected into these LLMs before their deployment and adoption. We propose UniVaR, a high-dimensional neural representation of symbolic human value distributions in LLMs, orthogonal to model architecture and training data. This is a continuous and scalable representation, self-supervised from the value-relevant output of 8 LLMs and evaluated on 15 open-source and commercial LLMs. Through UniVaR, we visualize and explore how LLMs prioritize different values in 25 languages and cultures, shedding light on complex interplay between human values and language modeling.

Effective planning requires strong world models, but high-level world models that can understand and reason about actions with semantic and temporal abstraction remain largely underdeveloped. We introduce the Vision Language World Model (VLWM), a foundation model trained for language-based world modeling on natural videos. Given visual observations, the VLWM first infers the overall goal achievements then predicts a trajectory composed of interleaved actions and world state changes. Those targets are extracted by iterative LLM Self-Refine conditioned on compressed future observations represented by Tree of Captions. The VLWM learns both an action policy and a dynamics model, which respectively facilitates reactive system-1 plan decoding and reflective system-2 planning via cost minimization. The cost evaluates the semantic distance between the hypothetical future states given by VLWM roll-outs and the expected goal state, and is measured by a critic model that we trained in a self-supervised manner. The VLWM achieves state-of-the-art Visual Planning for Assistance (VPA) performance on both benchmark evaluations and our proposed PlannerArena human evaluations, where system-2 improves the Elo score by +27

Large language models (LLMs) often generate responses that deviate from user input or training data, a phenomenon known as "hallucination." These hallucinations undermine user trust and hinder the adoption of generative AI systems. Addressing hallucinations is essential for the advancement of LLMs. This paper introduces a comprehensive hallucination benchmark, incorporating both new extrinsic and existing intrinsic evaluation tasks, built upon clear taxonomy of hallucination. A major challenge in benchmarking hallucinations is the lack of a unified framework due to inconsistent definitions and categorizations. We disentangle LLM hallucination from "factuality," proposing a clear taxonomy that distinguishes between extrinsic and intrinsic hallucinations, to promote consistency and facilitate research. Extrinsic hallucinations, where the generated content is not consistent with the training data, are increasingly important as LLMs evolve. Our benchmark includes dynamic test set generation to mitigate data leakage and ensure robustness against such leakage. We also analyze existing benchmarks, highlighting their limitations and saturation. The work aims to: (1) establish a clear taxonomy of hallucinations, (2) introduce new extrinsic hallucination tasks, with data that can be dynamically regenerated to prevent saturation by leakage, (3) provide a comprehensive analysis of existing benchmarks, distinguishing them from factuality evaluations.

Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale. Despite their potentially transformative impact, these new capabilities are as yet poorly characterized. In order to inform future research, prepare for disruptive new model capabilities, and ameliorate socially harmful effects, it is vital that we understand the present and near-future capabilities and limitations of language models. To address this challenge, we introduce the Beyond the Imitation Game benchmark (BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 450 authors across 132 institutions. Task topics are diverse, drawing problems from linguistics, childhood development, math, common-sense reasoning, biology, physics, social bias, software development, and beyond. BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models. We evaluate the behavior of OpenAI's GPT models, Google-internal dense transformer architectures, and Switch-style sparse transformers on BIG-bench, across model sizes spanning millions to hundreds of billions of parameters. In addition, a team of human expert raters performed all tasks in order to provide a strong baseline. Findings include: model performance and calibration both improve with scale, but are poor in absolute terms (and when compared with rater performance); performance is remarkably similar across model classes, though with benefits from sparsity; tasks that improve gradually and predictably commonly involve a large knowledge or memorization component, whereas tasks that exhibit "breakthrough" behavior at a critical scale often involve multiple steps or components, or brittle metrics; social bias typically increases with scale in settings with ambiguous context, but this can be improved with prompting.

This paper establishes a formal information-theoretic framework for image captioning, conceptualizing captions as compressed linguistic representations that selectively encode semantic units in images. Our framework posits that good image captions should balance three key aspects: informationally sufficient, minimally redundant, and readily comprehensible by humans. By formulating these aspects as quantitative measures with adjustable weights, our framework provides a flexible foundation for analyzing and optimizing image captioning systems across diverse task requirements. To demonstrate its applicability, we introduce the Pyramid of Captions (PoCa) method, which generates enriched captions by integrating local and global visual information. We present both theoretical proof that PoCa improves caption quality under certain assumptions, and empirical validation of its effectiveness across various image captioning models and datasets.

Large language models (LLMs) are prone to hallucination stemming from misaligned self-awareness, particularly when processing queries exceeding their knowledge boundaries. While existing mitigation strategies employ uncertainty estimation or query rejection mechanisms, they suffer from computational efficiency and sacrificed helpfulness. To address these issues, we propose the Explicit Knowledge Boundary Modeling (EKBM) framework, integrating fast and slow reasoning systems to harmonize reliability and usability. The framework first employs a fast-thinking model to generate confidence-labeled responses, enabling immediate utilization of high-confidence outputs, whereas uncertain predictions trigger a slow refinement model for accuracy improvement. To align model behavior with our proposed object, we propose a hybrid training pipeline, enhancing self-awareness without degrading task performance. Evaluations on dialogue state tracking tasks demonstrate that EKBM achieves superior model reliability over uncertainty-based baselines. Further analysis reveals that refinement substantially boosts accuracy while maintaining low computational overhead. The framework establishes a scalable paradigm for deploying reliable LLMs in error-sensitive applications, effectively balancing accuracy and practical utility.

This paper describes our research on AI agents embodied in visual, virtual or physical forms, enabling them to interact with both users and their environments. These agents, which include virtual avatars, wearable devices, and robots, are designed to perceive, learn and act within their surroundings, which makes them more similar to how humans learn and interact with the environments as compared to disembodied agents. We propose that the development of world models is central to reasoning and planning of embodied AI agents, allowing these agents to understand and predict their environment, to understand user intentions and social contexts, thereby enhancing their ability to perform complex tasks autonomously. World modeling encompasses the integration of multimodal perception, planning through reasoning for action and control, and memory to create a comprehensive understanding of the physical world. Beyond the physical world, we also propose to learn the mental world model of users to enable better human-agent collaboration.

Patch-based image tokenization ignores the morphology of the visual world, limiting effective and efficient learning of image understanding. Inspired by subword tokenization, we introduce subobject-level adaptive token segmentation and explore several approaches, including superpixel, SAM, and a proposed Efficient and PanOptiC (EPOC) image tokenizer. Our EPOC combines boundary detection--a simple task that can be handled well by a compact model--with watershed segmentation, which inherently guarantees no pixels are left unsegmented. Intrinsic evaluations across 5 datasets demonstrate that EPOC's segmentation aligns well with human annotations of both object- and part-level visual morphology, producing more monosemantic tokens and offering substantial efficiency advantages. For extrinsic evaluation, we designed a token embedding that handles arbitrary-shaped tokens, and trained VLMs with different tokenizers on 4 datasets of object recognition and detailed captioning. The results reveal that subobject tokenization enables faster convergence and better generalization while using fewer visual tokens.

Large Language Models often generate factually incorrect but plausible outputs, known as hallucinations. We identify a more insidious phenomenon, LLM delusion, defined as high belief hallucinations, incorrect outputs with abnormally high confidence, making them harder to detect and mitigate. Unlike ordinary hallucinations, delusions persist with low uncertainty, posing significant challenges to model reliability. Through empirical analysis across different model families and sizes on several Question Answering tasks, we show that delusions are prevalent and distinct from hallucinations. LLMs exhibit lower honesty with delusions, which are harder to override via finetuning or self reflection. We link delusion formation with training dynamics and dataset noise and explore mitigation strategies such as retrieval augmented generation and multi agent debating to mitigate delusions. By systematically investigating the nature, prevalence, and mitigation of LLM delusions, our study provides insights into the underlying causes of this phenomenon and outlines future directions for improving model reliability.

LLMs often adopt an assertive language style also when making false claims. Such ”overconfident hallucinations” mislead users and erode trust. Achieving the ability to express in language the actual degree of uncertainty around a claim is therefore of great importance. We find that ”verbal uncertainty” is governed by a single linear feature in the representation space of LLMs, and shows that this has only moderate correlation with the actual ”semantic uncertainty” of the model. We apply this insight and show that (1) the mismatch between semantic and verbal uncertainty is a better predictor of hallucinations than semantic uncertainty alone and (2) we can intervene on verbal uncertainty at inference time and reduce confident hallucinations on short-form answers, achieving an average relative reduction of ~30%.

In-context learning (ICL) empowers large language models (LLMs) to perform diverse tasks in underrepresented languages using only short in-context information, offering a crucial avenue for narrowing the gap between high-resource and low-resource languages. Nonetheless, there is only a handful of works explored ICL for low-resource languages with most of them focusing on relatively high-resource languages, such as French and Spanish. In this work, we extensively study ICL and its cross-lingual variation (X-ICL) on 25 low-resource and 7 relatively higher-resource languages. Our study not only assesses the effectiveness of ICL with LLMs in low-resource languages but also identifies the shortcomings of in-context label alignment, and introduces a more effective alternative: query alignment. Moreover, we provide valuable insights into various facets of ICL for low-resource languages. Our study concludes the significance of few-shot in-context information on enhancing the low-resource understanding quality of LLMs through semantically relevant information by closing the language gap in the target language and aligning the semantics between the targeted low-resource and the high-resource language that the model is proficient in. Our work highlights the importance of advancing ICL research, particularly for low-resource languages.

We propose to measure political bias in LLMs by analyzing both the content and style of their generated content regarding political issues. Existing benchmarks and measures focus on gender and racial biases. However, political bias exists in LLMs and can lead to polarization and other harms in downstream applications. In order to provide transparency to users, we advocate that there should be fine-grained and explainable measures of political biases generated by LLMs. Our proposed measure looks at different political issues such as reproductive rights and climate change, at both the content (the substance of the generation) and the style (the lexical polarity) of such bias. We measured the political bias in eleven open-sourced LLMs and showed that our proposed framework is easily scalable to other topics and is explainable.