All entries

Lessons learned while building Apex-2

model-releasesr-localllama

Hi everyone, thank you so much for all the interest in my model. It's more than I expected. Here is a short summary of the trial and error I went through while building Apex-2. 1. GPUs were always the

Let's go https://huggingface.co/reflection!

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Let's go https://huggingface.co/reflection! Introducing Beam: a highly efficient agentic open model with 501B total parameters and 23B active. - Frontier reasoning efficiency - Advances the Western op

agentsclem-delangue--x

5 Oct 2026

Lexicographic Multi-Objective On-Policy Distillation

arXiv:2610.02359v1 Announce Type: cross Abstract: Reinforcement learning from verifiable rewards (RLVR) usually optimizes answer correctness, yet useful language-model behavior also requires high-qual

LiBRA: Detection-Aware Image Watermark Removal via Bidirectional Latent Optimization

arXiv:2610.03166v1 Announce Type: cross Abstract: Digital watermarking supports source attribution for AI-generated images, but its reliability depends on resistance to removal attacks. Some attacks a

Applications

Light Entropic Optimal Transport on Riemannian Manifolds

arXiv:2610.03085v1 Announce Type: new Abstract: Entropic Optimal Transport (EOT) has become a practical framework for learning stochastic couplings between complex distributions, with applications in

applicationsarxiv-cs-lg

Hardware

Lightweight and Resource-Efficient Perception for Robotic Guide Dogs

arXiv:2610.03187v1 Announce Type: cross Abstract: Multi-camera streaming perception is increasingly deployed on heterogeneous edge platforms shared with co-resident workloads, yet accelerator placemen

hardwarearxiv-cs-cv

Lightweight, Rubric-Guided Trajectory Evaluation for Production AI Agents

arXiv:2610.03315v1 Announce Type: new Abstract: Trajectory evaluation is essential for improving the reliability of LLM-based agents, but production use makes it expensive to run repeatedly. Modern ag

Local Ai

LiteEMG-FM: An Efficient and Deployable Foundation Model for Robust EMG Sensing

arXiv:2610.02497v1 Announce Type: new Abstract: Electromyography (EMG) signals vary substantially across individuals, body regions, recording sessions, and sensing hardware, limiting the generalizatio

local-aiarxiv-cs-lg

LLA-MPC on Embedded Hardware: Rapid Adaptive Control with Thousands of Parallel Models

arXiv:2610.03616v1 Announce Type: new Abstract: We present a generalized implementation of Look-Back and Look-Ahead Adaptive Model Predictive Control (LLA-MPC), a learning-free framework for real-time

agentsarxiv-cs-ro

LLM-Guided Reinforcement Learning with Representative Agents for Traffic Modeling

arXiv:2511.06260v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly used as behavioral proxies for self-interested travelers in agent-based traffic models. Although

Localized Conformal Safety Monitoring with Vision-Language Models for Autonomous Driving

arXiv:2610.02765v1 Announce Type: cross Abstract: Monitoring planned driving trajectories requires accurately estimating the collision likelihood with actors whose motion is itself impacted by the ego

Applications

LOCUS: Landmark-Oriented Container Discrimination Using Spatial Graphs

arXiv:2610.02803v1 Announce Type: new Abstract: As robots are increasingly deployed in unstructured, real-world environments, the ability to reason about complex spatial and semantic relationships amo

applicationsarxiv-cs-ro

LoGo: Local-Global Rewards for Consistent Long-Horizon Video Generation

arXiv:2610.03636v1 Announce Type: cross Abstract: Camera-controlled video models are rapidly advancing toward long generation horizons and complex camera control. A key failure mode is 3D inconsistenc

Lost in the Request: How Communication Variation Disrupts Retrieval and Action in Email Agents

arXiv:2610.02627v1 Announce Type: new Abstract: An email assistant should not complete less work simply because a user phrases the same request differently. Yet most benchmarks test each task with onl

Low-Cost Video--Time Priors as a Strong Baseline for EEG--fNIRS Emotion Regression on Familiar Videos

arXiv:2610.03618v1 Announce Type: new Abstract: Continuous emotion regression estimates moment-to-moment valence and arousal while a viewer watches a video. In familiar-video deployment, responses fro

LPS-Bench: Benchmarking Safety Awareness of Computer-Use Agents in Long-Horizon Planning under Benign and Adversarial Scenarios

arXiv:2602.03255v2 Announce Type: replace Abstract: Computer-use agents (CUAs) execute multi-stage tasks through tools, where an early unsafe decision can propagate to consequential actions. Evaluatin

LS-AR: Future-Predictive Latent Steering in Autoregressive LLMs

arXiv:2610.03093v1 Announce Type: new Abstract: Standard autoregressive (AR) models process high-level task instructions, state history, and transient tokens within a single shared sequence of tokens.

LUMOS: Tracing Parametric Knowledge from Training Data to Behavioral Outputs in LLMs

arXiv:2610.02902v1 Announce Type: new Abstract: Current analyses of LLMs' parametric knowledge are largely output-centric, drawing conclusions about what a model knows without verifying what it was ac

MACTS-EM: Multi-Agent Collaborative Time Series Forecasting with Emergent Memory

arXiv:2610.02255v1 Announce Type: new Abstract: Time series forecasting remains a critical challenge across numerous domains. Despite significant advancements, existing approaches struggle with comple

agentsarxiv-cs-lg

Making Amazon Quick enterprise-ready: Automated, auditable cross-account resource promotion

Promoting Amazon Quick resources (agents, action connectors, knowledge bases, flows, and spaces) from a development to a production AWS account has been a manual, error-prone chore. This post shows ho

agentsaws-ml-blog

ManifoldSplat: Language-Guided Semantic Shape Editing of 3D Gaussian Head Avatars

arXiv:2610.03599v1 Announce Type: new Abstract: High-fidelity 3D head avatars have reached near-photorealistic quality. While recent methods enable text-driven manipulation, they struggle to provide f

safetyarxiv-cs-cv

ManiPhysicsBench: Physics-Based Assessment of Object Preservation in VLA Manipulation

arXiv:2610.02802v1 Announce Type: new Abstract: Vision-language-action (VLA) models aim to perform diverse manipulation tasks, but task success in existing rigid-body benchmarks does not indicate whet

researcharxiv-cs-ro

Applications

Mapping and Advancing the Scalability-Accuracy Frontier of Nonlinear Causal Discovery

arXiv:2610.03258v1 Announce Type: cross Abstract: Scalable nonlinear causal discovery requires methods that combine flexible mechanism estimators with efficient search over large graph spaces. Several

applicationsarxiv-cs-ai

Mastering Atari 2600 Games with Discovered Options

arXiv:2610.03604v1 Announce Type: new Abstract: Temporal abstractions, often instantiated as options, have long been regarded as a mechanism for accelerating credit assignment, facilitating exploratio

agentsarxiv-cs-lg

MEA: A Reward-Driven Multi-Agent System for Faithful Model Explanations

arXiv:2610.02480v1 Announce Type: new Abstract: Recent years have seen the employment of a plethora of machine learning (ML) models in high-stakes domains, but they remain largely opaque to the practi

Measure Less, Know More: Self-Supervised Test-Time Feature Acquisition

arXiv:2610.03454v1 Announce Type: cross Abstract: Recent progress in multimodal, high-dimensional learning has enabled foundation models to process heterogeneous, large-scale data. However, at test ti

Mental-R1: Aligning LLM Reasoning for Mental Health Assessment

arXiv:2606.13176v2 Announce Type: replace Abstract: Mental health problems such as anxiety, depression, and suicide remain urgent global challenges, where timely and accurate assessment is critical fo

MeshQuery: Agentic Seam Planning for UV Parametrization

arXiv:2610.02507v1 Announce Type: new Abstract: We present MeshQuery, a training-free agentic approach to automatic UV unwrapping of production-grade quad meshes. A Vision-Language Model (VLM) plans a

agentsarxiv-cs-cv

MetaRubric: Learning to Reward for Rubric-Based Reinforcement Learning

arXiv:2610.02824v1 Announce Type: new Abstract: Rubric-based reinforcement learning extends reward-driven optimization to open-ended tasks by assigning partial credit to individual response requiremen

Metropolis-Hastings Dominates Importance Resampling for Policy Composition

arXiv:2610.03480v1 Announce Type: new Abstract: Post-training a large language model (LLM) often requires exploring trade-offs between multiple rewards, but retraining for each trade-off is expensive.

safetyarxiv-cs-lg

MiDShip: Multimodal Dataset of Ship Cargo Hold Structures for Engineering Design

arXiv:2610.02214v1 Announce Type: cross Abstract: Ship structures govern vessel strength, safety, and manufacturability, but their design must satisfy hundreds of classification society requirements,

Mind the Refinement Gap: When Safe High-Level Robot Plans Produce Unsafe Executions

arXiv:2610.02662v1 Announce Type: new Abstract: Language-enabled robot systems increasingly combine semantic-graph planning with temporal-logic safety monitors. We investigate a trace-completeness ass

safetyarxiv-cs-lg

MintEval: Do LLMs Implement the Trading Strategy You Asked For? A Behavioural-Equivalence Benchmark for Natural-Language-to-Strategy Code

model-releasesarxiv-cs-cl

arXiv:2610.03080v1 Announce Type: cross Abstract: Large language models are moving from producing trading signals to writing the code that executes them. The failure mode of the second role is silent:

MintFlow: Minimal Trajectory Intervention for Constrained Flow Matching

arXiv:2610.02260v1 Announce Type: new Abstract: Flow matching models excel at generative modeling, and many downstream applications require their samples to satisfy prescribed constraints, such as obs

MIRROR: Multipath Quorum Integrity for LLM Multi-Agent Communication

arXiv:2610.02349v1 Announce Type: cross Abstract: Inter-agent communication is central to Large Language Model Multi-Agent Systems (LLM-MAS), but it introduces an underexplored vulnerability: Agent-in

Tutorials

Misinformation Without Triggers: From Factual Answers to Downstream Decisions

arXiv:2610.02886v1 Announce Type: cross Abstract: Language models learn from web documents, some of them false, and false content can reach a model's answer to a factual question and the summaries and

tutorialsarxiv-cs-ai

Mitigating Convergence Collapse in Fixed-Target Anomaly Detectors via Kernel-Anchored Locality Regularization

arXiv:2610.02345v1 Announce Type: new Abstract: A family of tabular anomaly detectors trains a neural map toward a fixed target under squared-error loss and scores anomalies by the test-time residual;

researcharxiv-cs-lg

Mitigating Private Data Leakage in LLMs with Whiteout

arXiv:2610.02418v1 Announce Type: cross Abstract: Modern large language models (LLMs) are trained on massive, largely unfiltered datasets, including content scraped from nearly every accessible websit

Mitigating Social Sycophancy via Pluralistic Preference Optimization

arXiv:2610.02568v1 Announce Type: new Abstract: Personal advice, including relationship advice, now ranks among the most common uses of generative AI. But language models (LMs) exhibit sycophancy: the

Mixture-of-Experts for Cryptocurrency Order Execution: Training Stability, Tail Risk, and Failure Modes

arXiv:2610.03369v1 Announce Type: cross Abstract: Deep reinforcement-learning policies for order execution can vary substantially across training seeds, so apparent architectural gains may reflect fav

MixVLA: Adaptive Mixing of Non-Invariant Information for Generalizable Vision-Language-Action Models

arXiv:2610.02898v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have achieved remarkable advances in robotic manipulation, yet their zero-shot generalization under out-of-distribut

safetyarxiv-cs-ro

MLCommons Jailbreak Benchmark v1.0

arXiv:2610.02827v1 Announce Type: new Abstract: Modern AI systems are designed to refuse hazardous requests. A jailbreak is a prompt crafted to bypass those safeguards and elicit outputs that the syst

MobiAgent: Dual-Loop Recursive Policy Self-Improvement for Long-Horizon Mobile Manipulation

arXiv:2610.03476v1 Announce Type: cross Abstract: Long-horizon mobile manipulation presents significant challenges due to compounding execution errors and capacity interference between locomotion and

Modeling Shared and Individual Structure for Cross-Subject Continuous Affect Regression from EEG-fNIRS

arXiv:2610.02796v1 Announce Type: new Abstract: Continuous, second-by-second valence-arousal estimation from physiological signals is typically studied in a subject-dependent setting, where the model

MoE SSD streaming on a 64 GB Mac mini: GPU still waits 27% of decode on experts. Ideas?

model-releasesr-localllama

Looking for some assistance /ideation. I am running qwen flash next q4 in my Mac mini m5 64gb. QFN doesn’t fit so this is done by having as many experts hot in cache as possible and streaming in the r

MOF-VERIFY: A Failure-Aware Agentic Harness for MOF Hypothesis Verification

arXiv:2610.03056v1 Announce Type: new Abstract: Large language models are increasingly used as reasoning components in AI-driven materials Co-Scientists, yet the reliability of the resulting verificat

Morality is Contextual: Learning Interpretable Moral Contexts from Human Data with Probabilistic Clustering and Large Language Models

arXiv:2512.21439v2 Announce Type: replace-cross Abstract: A key question in current AI alignment research is how to make AI algorithms learn moral values. Because human morality is highly context-depe

Morpheus: A Morphology-Aware Neural Tokenizer and Word Embedder for Turkish

arXiv:2606.18717v2 Announce Type: replace-cross Abstract: Turkish is agglutinative: meaning is carried by morphemes, yet the subword tokenizers that drive modern language models split words by corpus