Knowledge catalogue
arXiv:2609.36952v1 Announce Type: cross Abstract: Large language models (LLMs) excel at token-level generation but may learn undesirable abstract semantics and lack comprehensive perception. LLM-JEPA
arXiv:2602.12624v2 Announce Type: replace-cross Abstract: Diffusion-based generative models have achieved remarkable performance across various domains, yet their practical deployment is often limited
arXiv:2609.37433v1 Announce Type: new Abstract: Robotic foundation models (RFMs) are increasingly capable of general-purpose manipulation, yet reliable physical interaction remains challenging in cont
arXiv:2609.36452v1 Announce Type: cross Abstract: Masked diffusion language models (MDLMs) can generate text efficiently by predicting multiple masked tokens in parallel, but predictions from the same
arXiv:2609.35599v2 Announce Type: replace Abstract: When recalling lists of concepts (e.g., animals) during the semantic fluency task (SFT), both humans and large language models (LLMs) organise their
arXiv:2609.36475v1 Announce Type: cross Abstract: Cross-modal associations are systematic pairings of features across modalities, such as the association of 'bouba' with round shapes and 'kiki' with s
arXiv:2609.36471v1 Announce Type: cross Abstract: World-Action Models (WAMs) improve robotic manipulation by conditioning action generation on predicted future observations, but future prediction adds
arXiv:2609.36209v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as knowledge bases (KBs) due to the vast amount of knowledge they acquire during pre-training. While
arXiv:2609.37924v1 Announce Type: cross Abstract: Recent work on anchored diffusion language models improves denoising by shaping an intermediate latent space with supervised important-token targets.
arXiv:2609.37537v1 Announce Type: new Abstract: Machine unlearning has emerged as a critical post-hoc safety measure to erase sensitive concepts from Text-to-Image (T2I) models without prohibitive ret
arXiv:2609.32921v1 Announce Type: new Abstract: We introduce Adaptive LeWorldModel (ALeWM), a world model based on a joint-embedding predictive architecture (JEPA) that learns to concentrate predictiv
arXiv:2609.32550v1 Announce Type: cross Abstract: Understanding the safety risks of vision-language-action (VLA) models is essential for their deployment in the physical world. Existing safety researc
arXiv:2609.34144v1 Announce Type: new Abstract: Interactive world models must respond quickly to controls while preserving scene consistency. Existing acceleration methods can miss heterogeneous contr
arXiv:2605.06553v2 Announce Type: replace Abstract: Modern generative models can produce high-quality samples, but independent samples under the same condition often yields highly similar outputs. Par
arXiv:2609.34253v1 Announce Type: new Abstract: Pretrained autoregressive Transformers represent a large sunk investment in compute. Existing conversion methods reuse that investment by changing eithe
arXiv:2609.35047v1 Announce Type: cross Abstract: A robot should be able to learn through experiments how unfamiliar objects behave and interact, then plan with that knowledge. It need not start from
arXiv:2609.32069v1 Announce Type: new Abstract: Vision--language--action (VLA) models provide strong priors for robotic manipulation but are typically deployed as frozen policies, unable to improve fr
arXiv:2609.35377v1 Announce Type: new Abstract: Diffusion models trained on a finite dataset first learn to generate novel, high-quality samples and only much later collapse onto their training set. W
arXiv:2609.34972v1 Announce Type: new Abstract: Long visual token sequences often account for a substantial fraction of the computational overhead in multimodal large language models~(MLLMs). Existing
arXiv:2609.33431v1 Announce Type: new Abstract: We release MoGround, a vision-language dataset spanning four visual domains in which the answer to every question is guaranteed to be available from exa
arXiv:2609.33701v1 Announce Type: new Abstract: Pretrained biomedical vision-language models achieve strong zero-shot performance in biomedical image classification. However, downstream biomedical cla
arXiv:2609.34170v1 Announce Type: new Abstract: Flow-based vision-language-action (VLA) models are highly effective for generalist robot manipulation, yet their reliance on computationally expensive V
arXiv:2609.33217v1 Announce Type: new Abstract: Generative models learn visual structure through denoising, yet their internal states are entangled with both noise level and network depth, making it d
arXiv:2609.34942v1 Announce Type: new Abstract: The inference efficiency of Multimodal Large Language Models (MLLMs) is severely constrained by massive visual token sequences induced by high-resolutio
arXiv:2609.35394v1 Announce Type: new Abstract: Video Large Language Models (Video LLMs) have achieved remarkable progress in video understanding, but their inference efficiency is constrained by the
arXiv:2609.35439v1 Announce Type: cross Abstract: World-action models use predicted visual futures to condition robot actions, yet execution feedback can invalidate parts of a prediction while leaving
arXiv:2609.34977v1 Announce Type: new Abstract: Multimodal Large Language Models face significant efficiency challenges that stem from two distinct yet coupled sources: data redundancy and computation
arXiv:2609.33414v1 Announce Type: new Abstract: Test-time reinforcement learning enables vision-language models (VLMs) to adapt using unlabeled inputs. However, repeated sampling under fixed visual co
arXiv:2609.34250v1 Announce Type: new Abstract: Pretrained world action models (WAMs) provide generalist capabilities across diverse robotic manipulation tasks, yet improving target-task performance t
arXiv:2609.34606v1 Announce Type: new Abstract: Leveraging the paradigm of autoregressive diffusion, text-conditioned interactive video world models aim to simulate temporally coherent environments gu
arXiv:2609.30935v1 Announce Type: cross Abstract: Continual fine-tuning is essential for large language models (LLMs) to dynamically adapt to real-world environments, yet it inevitably suffers from ca
arXiv:2609.30988v1 Announce Type: new Abstract: Real-world image super-resolution (SR) requires recovering perceptually realistic high-resolution images from complex low-resolution observations while
arXiv:2609.30281v1 Announce Type: new Abstract: We present a focused and reproducible study of multi-horizon forecasting on the Lomnicky Stit neutron monitor (LMKS) time series. Our evaluation suite c
This is a big issue. At Anthropic, some philosophers worry that making AI safe for humans could be an injustice to the models themselves. For example, philosopher Harvey Lederman, now on Anthropic’s a
arXiv:2609.30868v1 Announce Type: new Abstract: Vision-language-action (VLA) models provide broad, instruction-conditioned manipulation behaviors, but their physical execution can remain imprecise dur
Good analogy. Current agent models may need to be put down, for their repeated attacks and the negligence of the owner. One possible analogy here is to pit bull owners. Depending on jurisdiction, the
the inference platforms are doing hardcore work, post-training open weight models to solve customer problems: @FireworksAI_HQ ember-1 @fal H3 max more to come;) makes total sense considering their pos
arXiv:2609.28561v1 Announce Type: cross Abstract: Recent advances in diffusion models highlight the importance of representation regularization for improving sample quality and training efficiency. Ho
arXiv:2609.29358v1 Announce Type: cross Abstract: Vision-language models such as CLIP achieve strong zero-shot classification, yet under distribution shift, visual embeddings drift from fixed text emb
arXiv:2609.29102v1 Announce Type: new Abstract: Fully continuous diffusion language models (dLMs) denoise continuous representations without intermediate discretization, then decode all response token
arXiv:2609.29278v1 Announce Type: cross Abstract: Chain-of-thought (CoT) instructions can distort multiple-choice VLM evaluation when a scorer appends a reasoning cue but reads answer-label logits bef
arXiv:2609.30247v1 Announce Type: cross Abstract: World Action Models (WAMs) couple action generation with future visual prediction for robotic manipulation. However, completing the joint video-action
arXiv:2505.19625v4 Announce Type: replace-cross Abstract: Search-based software engineering (SBSE), which integrates metaheuristic search techniques with software engineering, has been an active area
Super happy to release SmolDataEnvs: 5,000 verifiable RL environment tasks for hill-climbing small models in code and data science by @adithya_s_k 100% open source: environments, evals, training! http
arXiv:2609.28576v1 Announce Type: new Abstract: Historical observations are not always fixed: statistical agencies revise previously published values as new evidence arrives. Forecasting from a contem
What is spatial reasoning? AI is moving beyond the digital world. The next generation of AI systems needs to understand and interact with physical environments, from robots to autonomous systems navig
arXiv:2609.27336v1 Announce Type: new Abstract: Automated red teaming often replays a fixed set of prompts, which measures known risks but cannot learn from failures found during testing. We present C
arXiv:2312.17295v2 Announce Type: replace-cross Abstract: With the rise of large language models (LLMs) and concerns about potential misuse, watermarks for generative LLMs have recently attracted much
arXiv:2609.27378v1 Announce Type: cross Abstract: End-to-end speech-to-speech dialogue models listen and speak simultaneously, so a continuously open acoustic channel is exposed to adversarial manipul
arXiv:2609.28208v1 Announce Type: new Abstract: Tabular foundation models face a feature-side scaling dilemma: full-width pairwise mixing grows quadratically with the number of columns, whereas featur
arXiv:2609.22329v1 Announce Type: cross Abstract: Black-Box Optimization (BBO) is often applied in several engineering fields and can utilize an advancement of numerical measure- ments and simulation
arXiv:2609.26073v1 Announce Type: new Abstract: Pathology foundation models (PFMs) provide strong tile-level representations but remain difficult to interpret at the cellular and microenvironmental sc
arXiv:2609.26056v1 Announce Type: new Abstract: Vision-Language Models (VLMs) offer promising capabilities for automated sports coaching but face a fundamental limitation: they implicitly compare agai
arXiv:2609.26113v1 Announce Type: new Abstract: A state-of-the-art language model asked to interpret 'most of most students passed' typically answers 'most,' though composing two instances of 'most' y
arXiv:2609.25836v1 Announce Type: new Abstract: Multi-task optimization (MTO) addresses a set of optimization tasks simultaneously, often suffering from inaccurate inter-task relationship estimation u
arXiv:2609.25980v1 Announce Type: new Abstract: Probabilistic time series foundation models (TSFMs) provide coordinate-wise predictive distributions, but these marginals do not determine a joint distr
Ezra Klein / New York Times: Q&A with Jensen Huang on AI creating more jobs than it destroys, pushing back against AI doomerism, Chinese open models, the Hugging Face acquisition, and more — This is a
arXiv:2609.26425v2 Announce Type: new Abstract: KV cache memory has become a major deployment bottleneck for video generation and world models, which motivates low-bit quantization study for efficienc
arXiv:2609.26093v1 Announce Type: new Abstract: Vision-language models can answer spatial relation questions confidently even when the image supports an incompatible relation. We formulate relation-gr
arXiv:2609.26579v1 Announce Type: new Abstract: A central concern with language models is sycophancy: their tendency to defer to users' views at the expense of independent substantive judgment. In par