All entries

Wasserstein Causal Forests for Distribution-Valued Outcomes

arXiv:2609.35898v1 Announce Type: cross Abstract: This paper proposes Wasserstein Causal Forests (WCF) for settings in which each unit's outcome is itself a probability distribution. This study also d

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Watch-Think-Interact: Bootstrapping Long-Horizon Multi-Turn Streaming Video Reasoning with Reinforcement Learning

DGX agent

arXiv:2609.37035v1 Announce Type: new Abstract: Streaming video assistance requires models to answer asynchronous questions from an observed prefix under a fixed context budget. Existing approaches mo

researcharxiv-cs-ai

30 Sept 2026

WayFinder: Hierarchical Visual-Language-Action for Zero-Shot Waypoint Generation and Low-Level Kinematic Control

arXiv:2609.37922v1 Announce Type: new Abstract: Visual Language Action (VLA) models offer unprecedented generalization for autonomous robots; however, their real-world deployment is frequently bottlen

safetyarxiv-cs-ro

Waypoint-1.5: A Real-Time Video World Model for Consumer Hardware

arXiv:2609.37107v1 Announce Type: new Abstract: We present Waypoint 1.5, a real-time diffusion world model for interactive video generation on consumer-grade hardware. Unlike general video diffusion m

safetyarxiv-cs-cv

Tools

We built a new way to train contextual embedding models, which encode each chunk of a document with the whole document in view. pplx-embed-v…

We built a new way to train contextual embedding models, which encode each chunk of a document with the whole document in view. pplx-embed-v2-context-9b-preview sets a new state of the art on ConTEB a

toolsperplexity--x

we comprehensively benchmarked gpt-6.1 sol on document OCR tasks. compared to gpt-6 sol from a week ago (!), we see a sizable increase in ta…

we comprehensively benchmarked gpt-6.1 sol on document OCR tasks. compared to gpt-6 sol from a week ago (!), we see a sizable increase in table parsing capabilities and reading order, which is pleasan

agentsjerry-liu--x

We just open-sourced the world's fastest WebGPU kernels for local AI on Hugging Face

The collection includes kernels for more than 200 common ML operations, all of which can run entirely locally in your browser on WebGPU. We're also working to upstream these optimizations to Transform

local-air-localllama

Weeding Out Bad Seeds: Initial-Noise-Robust Unlearning for Text-to-Image Diffusion Models

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

safetyarxiv-cs-cv

WEFT: Scaling Tool-Use Post-Training for General-Purpose Agents

arXiv:2609.36887v1 Announce Type: new Abstract: Recent efforts to scale tool-use post-training have largely centered on the synthesis of executable environments, which constitute only one component of

agentsarxiv-cs-ai

Weight-Adjusted Gradients Reveal Parameter Importance and Failure Modes in LLMs

arXiv:2607.10803v2 Announce Type: replace-cross Abstract: Understanding which parameters are influential in Large Language Models (LLMs) is central to improving their efficiency, reliability, and inte

Weights Read and Write Features: Scalable Parameter Decomposition Grounded in Activation Space

arXiv:2609.37731v1 Announce Type: new Abstract: Activation space and parameter space provide complementary views of model computation. Activations represent information, while weights read, transform,

WeLike2Party! In-Context Motion Transfer for Multi-Human Image Animation

arXiv:2609.36937v1 Announce Type: cross Abstract: Human image animation aims to transfer motion from a driving video to subjects in a reference image. Despite remarkable progress in video generation,

“We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer

Two months after the bombshell news that a swarm of its agents had broken their containment and hacked into the computers of the AI company Hugging Face, OpenAI is still putting out fires. A steady dr

agentsmit-tech-review

What AI model can I run??

Processor Intel(R) Core(TM) Ultra 5 226V (2.10 GHz) Installed RAM 16.0 GB (15.7 GB usable) Graphics card Intel(R) Arc(TM) 130V GPU (8GB) (128 MB) Storage 273 GB of 954 GB used System type 64-bit opera

local-air-ollama

What Comes Next? Omni-StoryBench for Evaluating Story-Grounded Omnimodal Generation

arXiv:2609.37317v1 Announce Type: new Abstract: Omnimodal evaluation should go beyond independent text, image, and speech production: individually plausible outputs may not express a coherent shared e

What Does Post-Training Change in Multilingual Reasoning?

arXiv:2609.37104v1 Announce Type: cross Abstract: Open-source reasoning models provide unequal access to reasoning capability across languages. When a model can solve a problem but cannot deliver a co

What if automating AI R&D triggers an intelligence explosion?

arXiv:2609.36054v1 Announce Type: cross Abstract: In contrast to even a year ago, AI systems now write most of the code inside the companies that build them. As more of the AI research and development

What Makes High-Magnification Knowledge Transferable? A Study of Cross-Resolution Distillation in Whole-Slide Imaging

arXiv:2609.36407v1 Announce Type: new Abstract: Cross-resolution knowledge distillation aims to improve low-magnification whole- slide analysis by transferring high-magnification representations, yet

researcharxiv-cs-cv

What Makes Recurrence Effective in Looped Language Models?

arXiv:2609.36636v1 Announce Type: cross Abstract: Looped language models (LoopLMs) increase computational depth through parameter sharing, offering a path to scale inference computation without adding

What Shared Prefixes Hide: Trajectory Dropout for On-Policy Distillation

arXiv:2609.33455v2 Announce Type: replace Abstract: On-policy distillation (OPD) trains a student model on its own trajectories using dense token-level feedback from a stronger teacher model. Since ea

local-aiarxiv-cs-ai

WHAT THE F*CK. SpaceXAI is using Grok Bot to build Grok Bot. Its agents design, code, test, and ship the product. In this 45-minute session,…

WHAT THE F*CK. SpaceXAI is using Grok Bot to build Grok Bot. Its agents design, code, test, and ship the product. In this 45-minute session, Peng Zheng and Lauren Tanshowed their internal setup across

agentselon-musk--x

What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views

arXiv:2609.36881v1 Announce Type: new Abstract: Causal foundation models (CFMs) pre-trained on data generated from various structural causal models (SCMs) have been proposed for estimating causal effe

When Can Prefixes Compile LoRA? Exact Resource-Capped Tests for Frozen Attention

arXiv:2609.36766v1 Announce Type: new Abstract: Can a fixed continuous prefix replace a given low-rank adapter while the attention head stays frozen? In this research, we show that the answer depends

researcharxiv-cs-lg

When do data mixtures improve scaling laws? Insights from high-dimensional regression

arXiv:2609.38011v1 Announce Type: new Abstract: Modern machine learning systems are trained on mixtures of data from different domains, and choosing the right mixture can substantially improve downstr

When Does Correction Become Repair? Mechanistic Auditing of Internal Interventions in Tool-Using LLMs

arXiv:2609.36138v1 Announce Type: new Abstract: Before invoking external tools, an agentic LLM must select among a K-way action space: executing a call, seeking clarification, answering directly, or d

When Is Coarse Supervision Worth It? Cost-Aware Learning under Unknown Aggregation

arXiv:2609.36704v1 Announce Type: new Abstract: Modern learning systems often acquire supervision at multiple resolutions, trading annotation cost against information content. We study cost-aware two-

When Models Don't Manipulate Manifolds: The Geometry of a Comparison Task

arXiv:2609.37680v1 Announce Type: cross Abstract: One of the current premises of mechanistic interpretability research is that detailed accounts of the geometry of neural network representations can t

When Semantics Matter: Reliability-Aware Semantic-Rhythm Control for Co-Speech Gesture Generation

arXiv:2609.36685v1 Announce Type: new Abstract: Co-speech gesture generation aims to synthesize natural gestures that are both temporally synchronized with speech and semantically consistent with the

When Should Agents Check External State? Budgeting Observations for Stored Intentions

arXiv:2609.37125v1 Announce Type: new Abstract: Prospective memory allows an agent to retain an intention tied to a future condition, but the stored intention does not reveal whether that condition cu

When Should LLMs Trust Their Own Revisions? A Risk-Aware Study of Intrinsic Self-Correction

arXiv:2609.35832v1 Announce Type: cross Abstract: Intrinsic self-correction asks a language model to revise its own answer without receiving new external evidence. A second pass can recover mistakes,

When Successful Memories Mislead Embodied Agents:Memory Adaption For Task-Conditioned Execution

arXiv:2609.35808v1 Announce Type: cross Abstract: Experience reuse can reduce repeated exploration in embodied agents, but a trajectory that succeeded previously may be unsuitable for the current exec

researcharxiv-cs-ai

When to Adapt: Multi-Signal Domain Shift Detection for Efficient Training-Free Adaptation in Open-Vocabulary Segmentation

arXiv:2609.37602v1 Announce Type: cross Abstract: Robust and reliable perception is essential for autonomous robots operating in real-world environments, particularly in long-term missions where envir

When to Retrieve, When to Stay: Uncertainty-Aware Temporal Evidence Allocation for Streaming Video-LLMs

arXiv:2609.37345v1 Announce Type: new Abstract: Streaming video understanding requires Video Large Language Models (Video-LLMs) to reason over continuous visual streams under causal constraints. As th

When Tools Silently Lie: Evaluating and Mitigating Blind Compliance in Tool-Augmented Data Agents

arXiv:2609.37153v1 Announce Type: new Abstract: Tool-augmented data agents rely on tool outputs for analytical decisions. Yet successful execution can return plausible but incorrect evidence, requirin

agentsarxiv-cs-ai

Tutorials

When Trees Are Not Enough: Learning Mixed-Topology Feature Graphs with Adaptive Graph Sparse Autoencoders

arXiv:2609.36294v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) expose interpretable features in large language model activations, yet existing structured SAEs impose single-parent trees

tutorialsarxiv-cs-cl

When Updating Stops Being Learning: Rethinking LLM Self-Evolution via learnable information gain

arXiv:2609.36535v1 Announce Type: new Abstract: Self-evolution lets large language models (LLMs) improve iteratively using their own generated data, but often suffers from self-evolution degeneration:

researcharxiv-cs-cl

When Upstream Messages Override Correct Answers: A Controlled Study of Multi-Agent LLM Collaboration

arXiv:2609.36855v1 Announce Type: new Abstract: Multi-agent LLM systems rely on message passing among specialized agents to accomplish complex tasks. However, an upstream agent may provide useful info

agentsarxiv-cs-ai

Tutorials

Where Does Randomness Matter in Neural Cellular Automata?

arXiv:2609.36797v1 Announce Type: new Abstract: Stochastic cell updates are often used throughout the life of a neural cellular automaton (NCA), from backpropagation through time to final rollout. Thi

tutorialsarxiv-cs-lg

Where Does Staleness Accumulate? Pool Aware Effective Staleness Control for Asynchronous RL in LLM Post-Training

arXiv:2609.36830v1 Announce Type: new Abstract: Fully asynchronous reinforcement learning (RL) improves resource utilization in large language model post-training by overlapping rollout generation wit

Where Predictive Supervision Goes Shapes What VLA Policies Learn

arXiv:2609.36645v1 Announce Type: cross Abstract: Future prediction is increasingly used to improve vision-language-action (VLA) policies, based on the premise that anticipating scene evolution encour

Where Privacy Belongs: Placement Diagnosis and Certified Selection for Private Counterfactual Explanations on Graphs

arXiv:2609.37667v1 Announce Type: new Abstract: Counterfactual explanations for graph neural networks (GNNs) find the minimal intervention that flips a node's prediction--but computing one requires re

Where Root Cause Analysis Fails: A Retrieval-Reranking Decomposition

arXiv:2609.36686v1 Announce Type: new Abstract: Identifying the root cause of an anomaly among hundreds of sensors is critical for preventing safety incidents and costly downtime in complex monitored

Where Should Physics Enter a Molecular Crystal Generator?

arXiv:2609.36398v1 Announce Type: cross Abstract: Generative models make molecular crystal structure prediction fast, but their samples still exhibit geometric and packing violations. Physics can be i

local-aiarxiv-cs-ai

Where the Model Changes Its Mind: Hindsight-Divergence Localization for Efficient Reinforcement Learning with Verifiable Rewards

arXiv:2609.36864v1 Announce Type: new Abstract: Group-relative methods for reinforcement learning with verifiable rewards (RLVR) learn from differences in rollout outcomes. Independently sampling comp

safetyarxiv-cs-lg

Which Attention Heads are like the Human Head? Not the Ones that Compute

arXiv:2609.37991v1 Announce Type: new Abstract: Brain-AI alignment is often interpreted as a sign that model and brain perform similar computations. Whether the aligned units are causally involved in

Which papyrus HTR is good enough? Character-error-rate tolerance of four papyrological tasks on Greek texts

arXiv:2609.37755v1 Announce Type: new Abstract: Purpose: Most Greek papyri remain unpublished and undigitised; a handwritten text recognition (HTR) pipeline that transcribes them automatically would l

Which uncesored model smaller than Qwen3.8 27B?

Hi, I’m using Qwen3.8 27B in my vulnerability scanner, it does scan JS files, tries to generate exploits etc. I’m using uncensored version, the model is like 21GB. Running it on Mac M4 Pro 64GB ram, i

model-releasesr-ollama

Who Warmed the Archives? LLMs Overestimate Historical Warmth

arXiv:2609.37499v1 Announce Type: new Abstract: Historical archives are an under-used source for extending the instrumental climate record backward in time, and LLMs offer a way to extract the indices