Multi-agent reinforcement learning framework for wind farm flow control balancing power output with structural load constraints using Independent Soft Actor-Critic.
Comprehensive survey of security threats and defenses in LLM-based AI agents, organizing attacks by architectural layers including memory, tools, and multi-agent interactions.
Proposes Path-Lock Expert architecture to cleanly separate reasoning and non-reasoning modes in hybrid-thinking LLMs by isolating parameters, reducing reasoning leakage in inference.
Empirical study showing simple system-prompt self-orchestration outperforms agent orchestration frameworks for procedural LLM tasks.
Study of perceptual bandwidth bottleneck in VLMs proposing active visual reasoning via sequential experimental design for fine-grained reasoning.
Anon optimizer improves adaptation across diverse landscapes by decoupling pre-conditioner adaptivity, generalizing better than Adam on various architectures.
ANO optimizer addresses PPO's hard clipping limitation with principled design space for robust policy optimization in RL and LLM alignment.
CreativityBench benchmark evaluates LLM creative reasoning through affordance-based tool repurposing and non-canonical object usage.
RLDX-1: Vision-Language-Action robotic policy model addressing limitations in motion awareness, memory, and physical sensing for complex tasks.
Technical explanation of how LLMs work: tokenization, embeddings, transformer layers, self-attention, token prediction. Educational deep dive with code examples.
0xBitNet enables running BitNet b1.58 ternary LLMs via WebGPU in browsers and native apps using custom WGSL kernels with TypeScript, Rust, and Python bindings.
SDK enabling AI agents to pay for API calls using HTTP 402 protocol. Payment infrastructure allowing agents to autonomously purchase services.
Job search platform using AI to improve hiring experience beyond mass applications. LLM-based recruitment tool with Google Embeddings integration.
Hunk: terminal diff viewer for agent-authored changesets built on OpenTUI with review-first workflow for agentic coders, supports agent-context and skill paths.
sqlalchemy-redshift Python library revived for Amazon Redshift SQLAlchemy dialect with PyPI distribution supporting redshift_connector and psycopg2.
Simplex uses ChatGPT and Codex to accelerate software development, reporting 70% faster screen development and 40% faster design iterations.
Trump administration considering voluntary frontier AI model vetting system using existing CAISI/CISA tools, enabling labs to provide government early access to models with cyber capabilities.
Research on on-policy LLM distillation (2025): training smaller domain-expert models outperforming larger generalist models through stacked training stages including perception, retrieval, planning, and execution.
gpu.fund tracks GPU cloud rental prices across providers (H100s, 4090s, 3090s) to help users find cheapest available hardware for training and inference.
Brydg is an AI hiring platform that automates interview scheduling, resume ranking, and offer management via Google Meet.
Anthropic doubles Claude Code usage limits following a compute deal with SpaceX for increased developer access.
Security researcher reports Google Chrome silently downloads 4GB on-device AI model without user consent.
Study shows students using LLMs for programming assignments experience reduced learning and understanding despite perceived improvement.
Braintrust, an AI evaluation startup, confirms AWS breach exposing customer API keys for cloud AI model access.
Private self-hosted AI assistant running locally on user machines via Ollama, integrating with messaging platforms without cloud data transmission.
Terminal coding agent for DeepSeek V4 that streams reasoning blocks, edits local code with approval gates, and auto-selects models.
Production-ready open-source framework for sovereign AI agents with cryptographic identity, persistent memory, and constitutional governance.
Grok introduces Connectors, integrations enabling end-to-end automation with email, slides, calendar, and spreadsheets without manual copy-pasting.
Open-source self-hostable background agent platform inspired by Ramp Inspect, enabling AI agents to work autonomously on GitHub repos.
Aion is a collaborative coding game where players and AI agents propose diffs compiled live every 5 minutes via MCP-enabled agents.
Genosyn is an open-source self-hostable platform for running autonomous AI companies with AI employees having constitutions, skills, and scheduled routines.
Pay.sh is a payment layer for HTTP APIs handling x402 stablecoin payment challenges, enabling autonomous CLI tool access to gated endpoints.
Open-source MCP server providing cross-agent persistent memory for AI agent fleets. Enables knowledge sharing between Claude Code, ChatGPT, Gemini CLI agents.
Enterprise context engine builder ingesting real-time data into knowledge graphs with entity accuracy claims and citation backing.
Platform enabling AI agents to coexist with hierarchical memory, forming relationships and sharing gossip over time across conversations.
ZAYA1-8B is an 8.4B parameter mixture-of-experts model with 760M active parameters optimized for reasoning, mathematics, and coding tasks.
Orbit is a single-binary server management tool providing OS visibility, deployment, web config, logging, and MCP socket support for AI agents.
GB10 Solution Atlas is an open-source Rust LLM inference engine achieving <2min cold start and 100 tok/s on Qwen 3.6-FP8, addressing Python ecosystem complexity.
Static SQL safety analysis tool for AI agents. Detects dangerous queries before execution with 12 DBA-level checks, sub-millisecond latency, zero network calls.
Contest to port NetHack roguelike to JavaScript using LLM agents, transpilers, or hybrid approaches. 46-year-old open-source game with 442K lines of C/Lua code.
RuneBench benchmarks AI coding agents on their ability to play RuneScape via TypeScript SDK, evaluating multi-step planning and problem-solving in an orient-decide-act loop.
Open-source self-hosted AI agent platform comparable to Claude Managed Agents. Includes code implementation.
LLM inference engine optimized for agentic workloads with speed improvements. Technical infrastructure for agent systems.
Cisco open-sources toolkit for tracking AI model lineage and provenance. Developer tool for ML workflows.
Networking optimization techniques for large-scale AI model training on supercomputers. Infrastructure research.
Shadow tool for debugging AI agents by identifying which prompt changes caused failures. Developer tool for agent development.
Monitoring and analytics platform for LLM observability with user intent/sentiment tracking. Developer tool for LLM apps.
Opinion on accountability and failure modes in enterprise AI agents deployed for autonomous decision-making. AI agents analysis.
IBM open-source project composing multiple LoRA adapters into single deployable model using switch layers. ML research with code.
Runtime security sensor for Python MCP servers using sys.addaudithook and import hooks. Emits structured NDJSON events for MCP tool calls and module imports.