All entries

applicationsclem-delangue--x

Over the last few months at @river_ai_inc we’ve developed what we believe is the best system in the world for building your own internal custom models, driven by 10 years of experience training top mo

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Research

Overcoming Data Imbalances with Expert Selection Biases

DGX agent

arXiv:2610.13370v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models have become the standard architecture for frontier LLMs. The multiple parallel experts in each transformer block theoret

researcharxiv-cs-cl

12 Oct 2026

PACT: Privileged Trace Co-Training for Multi-Turn Tool-Use Agents

arXiv:2606.16215v2 Announce Type: replace-cross Abstract: Multi-turn tool-use agents must reason, call tools, and adapt to observations across several interaction turns. Post-training such agents is c

agentsarxiv-cs-ai

PANDA: Prior-guided Attentional Dual-path Architecture

arXiv:2610.13635v1 Announce Type: new Abstract: Hypergraph Neural Networks (HNNs), designed to model higher-order relations beyond pairwise interactions, typically rely either on fixed structural weig

PanTact: Teaching Whole-Body Dexterity with a Wearable Contact Interface

arXiv:2610.12558v1 Announce Type: new Abstract: Humanoid robots have many degrees of freedom and contact surfaces, yet most loco-manipulation research restricts manipulation to the hands and avoids ot

safetyarxiv-cs-ro

Partially Informed Stochastic Optimization via Zeroth-Order Residual Tracking

arXiv:2610.13871v1 Announce Type: new Abstract: Many optimization problems provide direct access to only part of the gradient, while the remaining component must be inferred from function evaluations.

Local Ai

PartiGen: Part-Aware Representations for Language-Conditioned Humanoid Motion Generation

arXiv:2610.13373v1 Announce Type: new Abstract: Translating language into humanoid motion requires coordinating body parts, determining motion duration, and producing references suitable for robot exe

local-aiarxiv-cs-ro

Hardware

PATACON: GPU-Parallel Tangent-Bundle RRT-Connect for Constrained Motion Planning

arXiv:2610.13195v1 Announce Type: new Abstract: Fast whole-body motion planning enables continuous replanning for humanoid robots in dynamic environments while satisfying collision avoidance and kinem

hardwarearxiv-cs-ro

PDE-PFN: Prior-Data Fitted Neural PDE Solver

arXiv:2610.13334v1 Announce Type: new Abstract: Motivated by the success of large language models (LLMs) with broad generalizability and robustness to noisy or unreliable pre-training data, we explore

People Making Conversation: Evaluating How Naturally Speech Models Converse

arXiv:2610.12802v1 Announce Type: new Abstract: A conversational speech model has to decide, at every pause, whether to speak. Existing evaluations test boundary detection, policy adherence, or aggreg

Personalization Without Stereotyping: Do Speech LLMs Know When to Turn a Deaf Ear?

arXiv:2610.13550v1 Announce Type: new Abstract: Recent speech LLMs can leverage rich paralinguistic cues, enabling more complex and personalized interactions. However, little work investigates how thi

Personalized Federated Direct Preference Optimization via LoRA Adapters

arXiv:2610.12885v1 Announce Type: new Abstract: Federated Direct Preference Optimization (FedDPO) enables privacy-preserving alignment of language models across distributed agents without centralizing

perturber: A Stateless Service for Semantics-Preserving Text Perturbations

arXiv:2610.12667v1 Announce Type: new Abstract: Large language model (LLM) evaluations on clean benchmarks overstate model robustness: real inputs arrive perturbed, and these meaning-preserving varian

Phasis develops platform for agentic orientation using World APIs and reality layers

When the founders at Phasis Inc. sought to develop a platform for orienting AI agents, they turned to graphs as a way to obtain a measurement of culture. The company wanted to better understand relati

agentssiliconangle

Physics-inspired Generative AI models via real hardware-based noisy quantum diffusion

arXiv:2505.22193v2 Announce Type: replace-cross Abstract: Quantum Diffusion Models (QDMs) are an emerging paradigm in Generative AI that aims to use quantum properties to improve the performances of t

PlacePulse: Learning Transferable Place Representations from Structural Geographic Data

arXiv:2610.13390v1 Announce Type: new Abstract: Geospatial machine learning often depends on task-specific feature engineering over large collections of sociodemographic, business, and environmental v

Planning with Transformers: Chain of Computation and Structured Context Windows

arXiv:2607.17710v2 Announce Type: replace Abstract: Large Language Models (LLMs) have had a remarkable impact across many areas of machine learning. However, recent studies have shown that they strugg

Playing the network backward: A Game Theoretic Attribution Framework

arXiv:2605.06212v2 Announce Type: replace Abstract: Attribution methods explain which input features drive a model's prediction, making them central to model debugging and mechanistic interpretability

Polyconvex neural-network surrogates with learned paired directional features for anisotropic hyperelastic homogenization

arXiv:2610.12801v1 Announce Type: cross Abstract: Approximating the effective anisotropic hyperelastic response of a representative volume element (RVE) requires constitutive expressiveness, mechanica

PPG-Bench: A Unified Benchmark for PPG Foundation Models

arXiv:2610.13435v1 Announce Type: new Abstract: Photoplethysmography (PPG) is increasingly recorded by wearable and clinical devices at enormous scale, and a growing number of PPG foundation models (F

Prediction-Informed Landing Window Planning for UAV Landing on USV in High Sea States

arXiv:2610.13190v1 Announce Type: cross Abstract: Autonomous landing of unmanned aerial vehicle (UAV) on unmanned surface vehicle (USV) is difficult in high sea states. Existing hierarchical landing m

agentsarxiv-cs-lg

Tutorials

Prediction Is Not a Sufficient Memory State

arXiv:2610.13881v1 Announce Type: new Abstract: A memory that reproduces every current prediction may still have forgotten how to learn from the next observation. We characterize this distinction for

tutorialsarxiv-cs-lg

PreSTO: Predictive Subtree Prefetching for Fast LLM Power Sampling

arXiv:2610.13171v1 Announce Type: new Abstract: Power sampling improves large language model (LLM) reasoning at test time by sampling from a sharpened version of the base model's distribution. Its sam

researcharxiv-cs-cl

Pretraining and Test-Time Training Budgets Matter for ARC: A Controlled Study of VARC and TRM

arXiv:2610.13660v1 Announce Type: cross Abstract: The Abstraction and Reasoning Corpus (ARC) is increasingly used to evaluate whether neural networks can infer task-specific transformations from few d

Primal-Dual Flow Matching for Sample-Wise Constrained Generation

arXiv:2610.12699v1 Announce Type: new Abstract: We consider the problem of learning to generate data while satisfying a prescribed sample-wise constraint with a user-specified probability of constrain

PRIME: Pre-Retrieval Index Modeling Enhancement for Structured Document Retrieval-Real

arXiv:2610.13296v1 Announce Type: cross Abstract: High-precision retrieval over external knowledge bases is essential for LLM agents to operate effectively. However, existing Information Retrieval (IR

PriMoRetarget: Humanoid Motion Transfer Through Prioritized Motion Retargeting

arXiv:2610.12642v1 Announce Type: new Abstract: Motion retargeting is a critical preprocessing step for transferring human motion data to humanoid robots. While existing methods are effective for sing

researcharxiv-cs-ro

Private and Distributionally Robust Federated Multiclass Classification

arXiv:2610.13631v1 Announce Type: new Abstract: Fault diagnosis in asset-intensive industries requires multiclass models trained on noisy, sensitive data collected across geographically distributed cl

Proactive Agentic Orchestration

arXiv:2610.13485v1 Announce Type: new Abstract: Agentic orchestration delegates tasks to sub-agents with independent contexts, providing a key path to extending the effective execution horizon of larg

safetyarxiv-cs-cl

Probabilistic Forecasting of Electric Vehicle Charging in a Time-varying Hierarchical Infrastructure

arXiv:2610.13588v1 Announce Type: cross Abstract: Uncertainty quantification for electric vehicle (EV) charging demand is essential to turn the increasing pressure of EV adoption on power systems into

Local Ai

Probabilistic Ship Trajectory Forecasting using Transformer Models on AIS Data

arXiv:2610.13557v1 Announce Type: new Abstract: Anticipation of vessel trajectories is an essential task when navigating at sea. This applies to both human navigators and autonomous surface vessels. T

local-aiarxiv-cs-lg

Product Monte Carlo: Expectation Under a Product of Densities

arXiv:2610.12702v1 Announce Type: cross Abstract: Aggregating independently computed sub-sample inferences is a well-known challenge in large-scale Bayesian computation, federated learning, and opinio

ProgramGym: Scaling Training Environments for Long-Horizon Program Reproduction Agents

arXiv:2610.13395v1 Announce Type: cross Abstract: Coding agents are increasingly used to build programs from scratch as software engineers do rather than to resolve issues within a repository or to ge

agentsarxiv-cs-ai

Provably Accurate Shapley Value Estimation via Leverage Score Sampling

arXiv:2410.01917v3 Announce Type: replace-cross Abstract: Originally introduced in game theory, Shapley values have emerged as a central tool in explainable machine learning, where they are used to at

Provenance-Aware Belief Updating in Human-AI Feedback Loops

model-releasesarxiv-cs-ai

arXiv:2610.13008v1 Announce Type: new Abstract: In repeated human-AI interaction, a single observation can be quoted, summarized, and returned to a model as multiple reports, making repetition look li

Provenance-grouped protocol for evaluating machine learning in perovskite photovoltaics

arXiv:2610.12766v1 Announce Type: new Abstract: Machine-learning models for perovskite solar cells are typically evaluated using random data splits, which we show systematically leak information throu

Applications

PVBRF-ID: Private, Verifiable and Byzantine-Robust Federated Intrusion Detection for Oilfield Enterprise Networks

arXiv:2610.12704v1 Announce Type: cross Abstract: Federated learning allows dispersed oilfield sites to train a shared intrusion detector without pooling network traffic, yet its aggregation step forc

applicationsarxiv-cs-lg

Q&A with Mark Zuckerberg and Alexandr Wang on the Muse app, Muse Wonder, the risks of committing $100B+ to AI, rebooting Meta's AI team, agentic commerce, more (Cheeky Pint)

Cheeky Pint: Q&A with Mark Zuckerberg and Alexandr Wang on the Muse app, Muse Wonder, the risks of committing $100B+ to AI, rebooting Meta's AI team, agentic commerce, more — As Meta scales its AI inf

agentstechmeme

QLoRA on ExLlamaV3 Quants - Train @ 2-8bpw

EXL3-QLoRA I've built this over the last few months and I think it's stable enough to share now. A fork of ExLlamaV3, this is a native ExLlamaV3 training implementation with plain PyTorch, without tra

QTestBench: A Test Framework for Evaluating LLM-Generated Quantum Code

arXiv:2610.13811v1 Announce Type: cross Abstract: Large language models (LLMs) have shown strong results in classical code generation and are increasingly used to write the unit tests that safeguard t

Quality, Not Just Outcome: A Structural Scoring Framework for Selecting Coding-Agent Trajectories for Supervised Fine-Tuning

model-releasesarxiv-cs-ai

arXiv:2610.13152v1 Announce Type: new Abstract: Open coding agents are commonly trained through supervised fine-tuning on interaction trajectories whose final patches pass the benchmark test suite. Ho

Qwen 3.8 Flash Next Q3 S - Benchmark creation

After running a benchmark test on Exo in Opencode and being amazed by the result, I thought I'd try other models on Opencode. Many were surprisingly terrible. Especially some mid level models. Earlier

Qwen released QwenGyre - RL framework for training AI agents on extremely long complex tasks

Introducing QwenGyre, bringing elastic RL to extreme-long-horizon (XLong) agent training. It launches with two core designs: - Elastic Scheduler: reallocates GPUs between rollout & training as demand

Qwen3.8-27B vs Frontier vs Others

Here're results from my final test of coding, coding (panda/numpy), data analysis, data-driven decision-making, and local RAG domains I interact with most often. This test, which is refined from a 500

Qwen3.8-Flash-Next on six 3090s with vLLM: 7 users at 262k, 6.5k t/s prefill, 57 t/s decode solo / 206 t/s aggregate with all 7 generating. Fixes for the upstream included.

Write-up, serve command and raw data: https://github.com/lukolszewski/vllm This shows vLLM fixes and setup that let you put PLE on SSD and save 108GB of system RAM, use int8 kv and squeeze more users

Hardware

RAMP: Recurrent Adaptive Mixed Precision for Efficient Prefill in Recurrent-Depth Language Models

arXiv:2610.13041v1 Announce Type: new Abstract: Recurrent-depth language models improve inference-time reasoning by repeatedly applying a shared network to refine internal representations, but the add

hardwarearxiv-cs-cl

Randomized Momentum Transport with Cheap Predictors

arXiv:2610.13789v1 Announce Type: new Abstract: Momentum methods such as STORM and second-order momentum track the gradient by adding a transport correction to momentum, at the cost of an extra oracle

RARA: Robust Policy Fine-Tuning through Retrieval-Anchored Representation Alignment for Robotic Manipulation

arXiv:2610.13578v1 Announce Type: cross Abstract: Pre-trained robot policies, such as Vision-Language-Action (VLA) models, provide a strong foundation for generalist robotic manipulation, but adapting