Discover the Best AI Resources

Domain Arithmetic: One-Shot VLA Adaptation under Environmental Shifts

Adapts pretrained Vision-Language-Action (VLA) models to new camera poses and robot embodiments from a single demonstration by performing weight-vector arithmetic that injects domain-specific information. Filters noise via subspace alignment of singular components; designed for one-shot adaptation under visual and embodiment shifts.

Leanstral 1.5 119B A6B

A code-agent model for Lean 4 that automates repository-level formal proofs and verification; a Mixture-of-Experts architecture (119B total, 6.5B active) with 256k context, multimodal input and an Apache-2.0 license.

ELDR: Expert-Locality-Aware Decode Routing for PD-Disaggregated MoE Serving

Predicts per-request MoE expert footprints from prefill activations and routes decode requests to workers that maximize expert-locality, lowering decode latency by combining offline K-means partitioning with online locality-band routing and a KV-block–coindexed signature cache.

EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments

Provides a benchmark and protocol to evaluate agents that iteratively edit executable policies under a fixed interaction budget, recording full execution–feedback–revise trajectories. Built from compact RL environments with trajectory-level diagnostics and hidden held-out validation.

Multi-Resolution Flow Matching: Training-Free Diffusion Acceleration via Staged Sampling

Accelerates text-to-image diffusion for pretrained flow-matching models using a staged low-to-high-resolution pipeline: fast low-res sampling, pixel-space GAN super-resolution, light latent noising, and short high-res refinement — >10× end-to-end speedups without retraining.

Embodied.cpp: A Portable Inference Runtime of Embodied AI Models on Heterogeneous Robots

Provides a portable C++ inference runtime to deploy embodied AI models (vision–language–action and world–action) on heterogeneous robot hardware, enabling latency-first batch-1 closed-loop control. Key features include modular multi-rate layers, fused low-latency inference, and extensible head/IO plugins.

GigaWorld-1: A Roadmap to Build World Models for Robot Policy Evaluation

Provides a systematic benchmark and design roadmap for video-based world models to evaluate robot policies, introducing WMBench and GigaWorld-1 optimized for long-horizon, action-faithful rollouts. Offers controlled comparisons across model families, action encodings, and 324k+ simulated vs real rollouts, with code, models, and datasets released for reproducible evaluation.

grug-think

Provides 100,891 JSON-formatted agent conversation examples where each assistant turn includes a short <think> internal reasoning trace before tool/function calls. Human-facing text and tool calls are preserved; intended to fine-tune models to produce concise, cost-efficient chain-of-thought for tool use.

Program-as-Weights: A Programming Paradigm for Fuzzy Functions

Compiles natural-language function specifications into compact, locally-executable neural programs (PAW) that run on a small frozen interpreter; a 4B compiler emits LoRA adapters for a 0.6B runtime to provide offline, low-memory fuzzy text functions.

AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents

Introduces a bounded-memory, typed-retrieval contract for long-horizon LLM agents and evaluates it in Slay the Spire 2 — assembling per-decision prompts from five typed slots rather than appending raw transcripts. Key outputs include ablationable memory layers, 298 labeled trajectories, and reproducible analysis scripts.

Hy3

Provides a large Mixture-of-Experts instruct LLM (295B total parameters, 21B active, 256K context) optimized for reasoning, long-context retention and agent workflows; open-sourced under Apache-2.0.

VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon

Detects when an action-chunked VLA policy drifts from expected visual dynamics and triggers lightweight corrective replanning via a latent-space vision monitor and online gradient guidance; creates an event-driven adaptive action horizon without retraining the backbone.