Knowledge catalogue
arXiv:2605.09352v1 Announce Type: new Abstract: Understanding why independently trained neural networks from different modalities converge toward shared representations, and where this convergence lea
arXiv:2605.10130v1 Announce Type: new Abstract: Existing open-vocabulary detectors focus on RGB images and fail to generalize to thermal imagery, where low texture and emissivity variations challenge
arXiv:2605.10543v1 Announce Type: new Abstract: Director-style prompting, robotic action prediction, and interactive video agents demand temporal grounding over concurrent events -- a regime in which
tl;dr - let's not just remove negative behavior from models, but also add positive ones 👍 If anyone builds it, everyone thrives. Over the past decade, a lot of important work on AI alignment has focus
arXiv:2602.03688v2 Announce Type: replace Abstract: Multi-round LLM-based multi-agent systems rely on effective communication structures to support collaboration across rounds. However, most existing
arXiv:2605.09960v1 Announce Type: cross Abstract: Travel-time tomography forces a trade-off between mesh resolution and stability in which the regularizer choice dominates what can be recovered. We in
arXiv:2602.00678v4 Announce Type: replace Abstract: Reinforcement learning has shown strong promise for quadrupedal agile locomotion, even with proprioception-only sensing. In practice, however, sim-t
arXiv:2605.08129v1 Announce Type: new Abstract: Unified multimodal understanding and generation models enable richer human-AI interaction. Yet jointly customizing a character's persona, dialogue style
arXiv:2605.10194v1 Announce Type: new Abstract: On-policy self-distillation (self-OPD) densifies reinforcement learning with verifiable rewards (RLVR) by letting a policy teach itself under privileged
arXiv:2605.10843v1 Announce Type: cross Abstract: Large language models increasingly mediate decisions that turn on moral judgement, yet a growing body of evidence shows that their implicit preference
arXiv:2605.08741v1 Announce Type: new Abstract: Inference-time harnesses substantially improve large language models on complex reasoning tasks. However, the intrinsic capabilities of the underlying m
arXiv:2605.08511v1 Announce Type: new Abstract: Flow matching policies learn continuous velocity fields that transport noise to actions, enabling fast deterministic inference for robot manipulation. H
arXiv:2602.22779v2 Announce Type: replace Abstract: Tokenization in video models, typically through patchification, generates an excessive and redundant number of tokens. This severely limits video ef
arXiv:2511.03368v2 Announce Type: replace Abstract: The rise of the machine learning (ML) model economy has intertwined markets for training datasets and pre-trained models. However, most pricing appr
arXiv:2605.08964v1 Announce Type: new Abstract: In this thesis, we develop algorithms with theoretical guarantees for ensuring reliability and accountability of Machine Learning (ML) systems. As ML sy
arXiv:2605.09507v1 Announce Type: new Abstract: Video summarization aims to produce a compact representation of a long video by selecting a subset of temporally important segments that best reflect hu
arXiv:2511.12878v4 Announce Type: replace Abstract: Forecasting how human hands move in egocentric views is critical for applications like augmented reality and human-robot policy transfer. Recently,
arXiv:2605.10821v1 Announce Type: new Abstract: Diffusion-based vision-language-action (VLA) models have emerged as strong priors for robotic manipulation, yet adapting them to real-world distribution
arXiv:2603.05301v2 Announce Type: replace Abstract: Inductive spatio-temporal kriging infers signals at unobserved locations from observed sensors, but real-world observations are often incomplete and
arXiv:2605.08729v1 Announce Type: new Abstract: Motion, speech, and sound effects are fundamental elements of human-centric videos, yet their heterogeneous temporal characteristics make joint generati
arXiv:2605.10158v1 Announce Type: new Abstract: Process Reward Models (PRMs) are a powerful mechanism for steering large language model reasoning by providing fine-grained, step-level supervision. How
arXiv:2603.03971v2 Announce Type: replace-cross Abstract: Generative AI can convert uncertainty into hypersuasive, authoritative-seeming verdicts, displacing the justificatory work on which democratic
arXiv:2510.13830v2 Announce Type: replace-cross Abstract: Pairwise preference data have played an important role in the alignment of large language models (LLMs). Each sample of such data consists of
arXiv:2605.10172v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) have achieved remarkable success in general perception, yet complex multi-step visual reasoning remains a per
arXiv:2605.09920v1 Announce Type: cross Abstract: While Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as a promising post-training paradigm for Large Language Models (LLMs
arXiv:2504.04991v4 Announce Type: replace Abstract: Conventional visuomotor imitation learning usually predicts future robot actions directly in the time domain. Such formulations often have limited p
'We're going to reach a point of diminishing returns from scaling. That's what actually happened. The industry knows this. But doesn't want to tell you.' @GaryMarcus, AI Expert, Scientist & Author at
arXiv:2605.10716v1 Announce Type: new Abstract: Large language models are increasingly deployed with test-time strategies: sample N responses, score them with a reward model or verifier, and return th
arXiv:2602.02834v3 Announce Type: replace-cross Abstract: What structural inductive bias helps transformers reason over knowledge graphs? Through controlled ablations of a minimal transformer modifica
arXiv:2605.08828v1 Announce Type: new Abstract: Large language model agents increasingly operate through environment-facing scaffolds that expose files, web pages, APIs, and logs. These observations i
arXiv:2605.10659v1 Announce Type: cross Abstract: Digital personas powered by Large Language Models (LLMs) are increasingly proposed as substitutes for human survey respondents, yet it remains unclear
arXiv:2605.08245v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) increasingly power high-stakes applications, from medical imaging to autonomous systems, yet they routinely hallucinate,
arXiv:2605.08992v1 Announce Type: new Abstract: Federated learning (FL) is increasingly used to fine-tune foundation models (FMs) on distributed private data. The community largely assumes that large-
arXiv:2603.10051v2 Announce Type: replace-cross Abstract: Self-supervised masked modeling shows promise for encrypted traffic classification by masking and reconstructing raw bytes. Yet recent work re
Artificial intelligence guardrail and monitoring startup Pumpkin Intelligence Inc., which operates as White Circle, announced today it raised 11 million in seed funding from a who’s who of AI leadersh
arXiv:2601.21739v2 Announce Type: replace-cross Abstract: Adam has been at the core of large-scale training for almost a decade, yet a simple empirical fact remains unaccounted for: both validation sc
arXiv:2605.08250v1 Announce Type: cross Abstract: Recent advances in diffusion transformers (DiTs) have enabled promising single-turn image editing capabilities. However, multi-turn editing often lead
arXiv:2605.09765v1 Announce Type: cross Abstract: Representation learning in electronic health records (EHR) has largely followed paradigms inherited from natural language processing, relying on seque
wow! “It is the [House] Committee’s understanding that the new board of directors at OpenAI tried to address these problems upon your return by creating an “audit committee to review potential conflic
arXiv:2605.05611v2 Announce Type: replace-cross Abstract: In this paper, we present X-Voice, a 0.4B multilingual zero-shot voice cloning model that clones arbitrary voices and enables everyone to spea
arXiv:2605.10734v1 Announce Type: new Abstract: For reinforcement learning in the real world online exploration is expensive A common practice in robotic reinforcement learning is to incorporate addit
arXiv:2605.06866v1 Announce Type: new Abstract: Recent non-asymptotic analyses have substantially advanced the theory of distributional policy evaluation, but they largely concern synchronous full-sta
arXiv:2605.06938v1 Announce Type: cross Abstract: Building on the abstract Generalized Singular Value Decomposition (GSVD) theory of Brown et al. [2025], we prove that most modern neural architectures
arXiv:2602.13224v3 Announce Type: replace Abstract: Hallucinations in deployed language models can have real consequences for downstream decisions in domains such as healthcare, legal, and financial s
arXiv:2602.02320v3 Announce Type: replace-cross Abstract: Molecular function is largely determined by structure. Accurately aligning molecular structure with natural language is therefore essential fo
arXiv:2605.06749v1 Announce Type: cross Abstract: As learning systems increasingly shape everyday decisions, Algorithmic Collective Action (ACA), i.e., users coordinating changes to shared data to ste
arXiv:2605.08048v1 Announce Type: new Abstract: Measuring the breadth of a word's meaning, or its spread across contexts, has become feasible with contextualized token embeddings. A word type can be r
arXiv:2605.07857v1 Announce Type: new Abstract: Optimizing dynamic risk with stochastic policies is challenging in both policy updates and value learning. The former typically requires transition pert
arXiv:2605.07094v1 Announce Type: new Abstract: This paper introduces the Active-Importance-Sampling Actor-Critic (AISAC) algorithm, an extension of the Actor-Critic framework for reducing variance in
arXiv:2605.07257v1 Announce Type: new Abstract: Generative personalization often suffers from the semantic collapsing problem (SCP), where a learned personalized concept overpowers the rest of the tex
arXiv:2605.06717v1 Announce Type: cross Abstract: Coding agents are rapidly changing the landscape of software development, moving from inline completion to autonomous systems that edit repositories,
Ah yes, chatGPT, tell me more about the 'colony mind', 'foraging patrol' and 'nost building' behaviors of the human liver. For anyone predicting near-term physician replacement by LLM-based AI, please
AI existential crisis for software engineers is to go live in the woods and read poetry. AI existential crisis for creatives is to make things until 4am every day because you can't stop now. is this w
Am old enough to remember when @GeoffreyHinton told me I was stupid for saying that LLMs regurgitate training data. He was wrong. LLM regurgitation is now one of the best-established findings in the f
arXiv:2605.07825v1 Announce Type: cross Abstract: Training multimodal large language models has long been limited by the scarcity of high-quality paired multimodal data. Recent studies show that the s
arXiv:2605.07786v1 Announce Type: cross Abstract: As generative models achieve unprecedented visual quality, the gold standard for image evaluation remains traditional feature-distribution metrics (e.
arXiv:2601.18681v2 Announce Type: replace-cross Abstract: We consider time discretization for score-based diffusion models to generate samples from a learned reverse-time dynamic on a finite grid. Uni
arXiv:2604.01878v2 Announce Type: replace-cross Abstract: Spectral graph contrastive learning often constructs low- and high-frequency views to capture complementary graph signals, but these views are
arXiv:2605.06387v2 Announce Type: replace-cross Abstract: On-policy distillation (OPD) trains a student on its own trajectories with token-level teacher feedback and often outperforms off-policy disti
arXiv:2512.23694v2 Announce Type: replace-cross Abstract: Reliable long-horizon value prediction is difficult in offline reinforcement learning because fitted value methods combine bootstrapping, func