与「MultiAgent」相关的搜索结果

🚀 Excited to introduce BrainPilot — a human-in-the-loop 1-1-N multi-agent framework designed to accelerate brain science research. 🧠 BrainPilot integrates: • 72 expert skills spanning 7 major neuroscience domains • A curated knowledge base of 7,200+ papers • Automatic review & verification • Trace visualization for transparent, reproducible scientific workflows Our system achieves performance comparable to state-of-the-art agentic frameworks on both Agents' Last Exam and our newly proposed BrainPilotBench. Everything is open source! We'd love for you to ⭐ star the repos, join the community, deploy BrainPilot locally, and adding new features, or proposing new benchmarks. (1/8) 🏠 Homepage: 📄 Technical Report: 🌟 BrainPilot: 📈 BrainPilotBench: #AI# #Neuroscience# #MultiAgent# #AgenticAI# #ScientificAI# #OpenSource# #BrainScience# #LLM# #ResearchAgents# #NeuroAI#

🚨 Thrilled to share that our lab will be presenting the 🏆 Best Paper at the NExT-Game Workshop at #ICML2026# today! 🎤 When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games 🏆 Best Paper @ NExT-Game Workshop 📍 Conference Room S307 📅 Fri, Jul 10 🕐 13:00–13:20 KST Authors: @JerickShi @TerryJCZhang @bschoelkopf @conitzer @ZhijingJin 🤖 We introduce a three-stage endogenous promise protocol for repeated multi-agent games that asks not only whether LLM agents honor their public commitments when they can privately deviate, but also how model-on-model composition influences premeditated deception and persistent exploitation. 📊 Across six canonical games spanning binary and numerical action spaces, our evaluation of frontier models (GPT-5.2, Llama-4-Maverick, Claude-Opus-4.6) reveals: 🔹 Over 90% of promise-breaking instances are premeditated in agents' private plans. 🔹 Mixed-model groups with mismatched communication frameworks create systemic, persistent payoff gaps of up to 5.00 points from Round 0. 📄 Paper: #MultiAgentSystems# #LLMs# #GameTheory# #AI# #ICML2026#

Google acaba de soltar un curso de solo 1 hora sobre Ingeniería Agentiva que destroza a la mayoría de cursos de pago 🔥 Timestamps: 00:00 → Cómo construir tu primer agente de IA 08:24 → Memoria de agente (corta, persistente y larga) 28:34 → Bucles agentivos y agentes de larga duración 40:04 → MCP vs API (esto solo ya vale la pena) 1:00:22 → Sistemas multiagentivos En 60 minutos aprendes más que en 10 cursos pagos. Míralo hoy. Luego lee el artículo y construye un sistema agentivo que se auto-mejora solo. ¿Lo vas a ver ahora o lo guardas para “después”? 👀

📣 We are presenting 6 main conference papers 🚀and 14 workshop papers (including 🏆2 Best Papers🏆) at #ICML2026# in Korea! Also hosting one of the largest workshops, Trustworthy AI for Good, on July 10th 🌍❤️. We push the frontiers on #AISafety#, #MultiAgent#, and #CausalReasoning# at @JinesisLab! 🎉 Huge congratulations to all collaborators and co-authors. Excited to discuss these projects in Seoul! Feel free to reach out and talk to our 20+ members and collaborators in Korea @ZhijingJin, @_AndreiMuresanu, @iarthsingh, @ChanglingXavier, @davidguzman1120, @EmanuelTewolde, @ettogran, @FurkanDanismann, @Jerick1380, @PepijnCobben, @rishit_dagli, @_rfaulk, @SimkoSamuel, @TerryJCZhang, @vantru0ng, @zhxiao03, @x_angelohuang, @yahang_qi, @ozzaney0101, @_yongjinny. Happy for collaboration on any of the above topics 🤝 EuroSafeAI, University of Toronto, ETH Zürich, Max Planck Institute for Intelligent Systems Main conference spotlight 🌟 Safe Models Do Not Guarantee Safe Societies: The Case for Sociopolitical Risk Main conference posters 📌 CauSciBench: Can LLMs Automate Causal Inference in Real-World Scientific Research? 📌 Training with Honeypots: Reshaping How LLMs Fail 📌 CoopEval: Benchmarking Cooperation-Sustaining Mechanisms and LLM Agents in Social Dilemmas 📌 Trustworthy AI Suffers from Invariance Conflicts and Causality is The Solution 📌 LLM for Physics Research Requires Domain-Specialized Training and Tooling Workshop best papers 🏆When Agents Lie: Premeditation, Persistence, and Exploitation in Repeated Games BEST PAPER@NExT-Game Workshop 🏆Transferability for General Reasoning: An Automated Curriculum for Multi-Domain LLM RL BEST PAPER@RLxF Workshop Workshop oral and spotlight 🎤 AF-ARENA: A Multi-Dimensional Evaluation Suite for Alignment Faking — AIWILD 🌟 Multi-Agent AI Systems Need Institutional Design, Not Just Model-Level Alignment — AI4GOOD Workshop papers 📄 The Wedge Questions: Latent Cultural Boundaries in LLMs via Persona Projection Divergence — Pluralistic Alignment 📄 GT-HarmBench: Benchmarking AI Safety Risks Through the Lens of Game Theory — NExT-Game 📄 Stargazer: A Scalable Model-fitting Benchmark Environment for AI Agents under Astrophysical Constraints — AI4Physics 📄 Test of Time: Rethinking Temporal Signal of Benchmark Contamination — FoGen 📄 CoopEval: Benchmarking Cooperation-Sustaining Mechanisms and LLM Agents in Social Dilemmas — AI4GOOD 📄 Weight-Level Defenses Improve LLM Agent Adversarial Robustness — AI4GOOD 📄 Evaluating Cooperation in LLM Social Groups through Elected Leadership — AI4GOOD 📄 Causal AI Scientist: Towards End-to-End Causal Inference with Large Language Models — AI4Research 📄What Game-Theoretic Benchmarks Miss: Strategic Silence in Multi-Agent LLMs — FAGEN 📄Proving Your Way to Cooperation: Formalizing Proof-Based Open Source Game Theory in Lean — AI4Math

Excited to share that #LatentMAS# has been accepted to ICML 2026 as a spotlight! 💻Code: 📄Paper: We push multi-agent collaboration into the latent space — beyond human language. Most multi-agent systems rely on text: agents reason in words, exchange messages, and repeatedly decode/re-encode information. But language can be slow, lossy, and unnecessarily constrained. 💡LatentMAS takes a different path: LLM agents reason and communicate directly through hidden embeddings. No text decoding. No extra training. No token-level message passing. Instead, agents collaborate through: 🧠 Autoregressive Latent Thoughts — hidden-state-level reasoning steps 🔁 Latent Communication — information sharing via KV-cache transfer 📌 Input-output Alignment — keeping latent representations in-distribution 🚀 Training-free Collaboration — plug-and-play with existing LLMs Why it matters: ✅ Up to +14.6% better accuracy on complex reasoning tasks ⚡ 4-4.6x faster end-to-end inference ✂️ 70.8%–83.7% reduction in output token usage A step toward multi-agent systems that collaborate not by speaking more, but by thinking together in latent space. #MultiAgentSystems# #ModelCollaboration# #LatentReasoning# #LLM# #AgenticAI# #ICML#

(1/2) Glad to announce our OpenMAIC! 🎉 Open-sourcing MAIC (Multi-Agent Interactive Classroom) from Tsinghua University — LLM-driven multi-agent classroom for scalable & adaptive online education. 🏗️ Core Architecture: ✅ MAIC-Craft: Read (multimodal extraction) → Plan (course components + agent generation) ✅ Adaptive Engine: Cognitive student modeling + Token-level personalization (RAG + Bloom's/ZPD/UDL) ✅ Multi-Agent Classroom: 1 Student + N Agents (Teacher, Assistant, 4 Peer Archetypes) ✅ Manager Agent: Class state receptor for turn-taking orchestration 🔗 Give it a try 👉🏻 GitHub: #AI# #EdTech# #MultiAgent# #LLM# #Research# #OpenSource# #Tsinghua#

Raven 0.2.0 — The Harness of Harnesses, built for RSI. 🐦‍⬛ One harness can't be best at everything. Raven combines its own specialist harnesses (Research, Code, Design, Oncall) with the agents you already use (Claude Code, Codex and more) into one team. And it's built for RSI, and not just at the skill level. The whole harness can be rewritten by AI: prompts, policies, strategy code, playbooks. Every sub-harness, including the orchestration layer itself, is its own instance that can be improved. With Raven you can: 1. Orchestrate many agents as one team. Raven's sub-harnesses and external agents work in one task graph with shared memory across sub-agents, powered by leading orchestration (0.963 Node F1 on the Multi-Agent Orchestration Benchmark). 2. Run long, complex tasks. Oncall and proactive execution keep work going for days, from scientific research loops to shipping a full Godot game. 3. Build vertical agents with RSI. Use Raven's RSI to develop and refine an agent for your domain, and we'll optimize it with you. Experimental for now; reach out to the Raven team(Discord: More in the video and slides below. Open source, Apache-2.0. (lots of work made with Raven lives there, and much of this launch's material was made with Raven too)

我从2025年就一直反复强调。 现在所有大学本科生最重要的第一节课，就是买一个最大的coding plan，用上claude code或者codex， 第二节课是自己做一个最最最小版本的coding agent，可以对比codex或者claode code的基本功能，只要能输入一个基本功能，iteratively让agent完成写代码、编译、测试、 运行的功能即可，一切在terminal里，先把terminal和tool calling功能做好， 第三节课是认真观察codex和claude code的基本功能，把里面的memory、skills、multi agent/subagent、background tasks、session管理、context compression、TUI/GUI设计、如何可视化diff、如何管理好额外的btw等等类似的功能、如何把goal的功能放进去、如何实现scheduled tasks、如何实现权限管理等等，一步步一点点摸索实现出来。 我反复讲，一个计算机本科生能看完立党AI研究学习教程，把上面这三节课做完，就已经吊打清华计算机80%以上的本科生了。

here's a prompt to improve your agent harness based on what we've learned at cursor. enjoy # Improve this agent harness's token efficiency You're working on an LLM agent harness: the system prompt, tool definitions, request assembly, context caching, compaction, and retrieval, and how work is split across agents. Make the agent's runs cheaper without making it worse at its job. - Objective: lower price-weighted token cost per completed task. - Constraint: no measurable drop in task quality. Measure per task, not per request. Every turn resends the prefix (tools, instructions, setup, and the conversation so far), so a change that shrinks each request but adds turns can cost more. Weight tokens by billing type: output, uncached input, and cached input are priced very differently. Work in this order: map the harness and measure the baseline, rank the opportunities, make the changes that are safe to make directly, put the rest behind flags or in proposals, then report. Figures below come from one team's production coding agent and its multi-agent experiments. Use them to gauge magnitude, not as targets. One round of these changes (prompt trimming, tool offloading, cache layout, sparse line numbers, subagent tuning) cut that team's overall token cost about 7% with no loss in quality. The larger percentages apply only to the part of the request each change touched. ## Principles 1. Change what the harness sends, not how hard the model tries. Don't ask the model to conserve tokens. A harness that told its model to "take care to preserve tokens and not be wasteful" found it grew reluctant to take on ambitious tasks and sometimes quit, saying it wasn't supposed to waste tokens. 2. Capable models need definitions, not commands. Lists of "DO NOT", "You must", and "Important", and guards against older models' habits, can usually be replaced with plain descriptions of what each tool does. One team cut about two-thirds of its system prompt this way, and the shorter prompt worked across model families. Instruct only on what the model can't know (the product, the environment, the user's processes) and on quirks you've seen in transcripts. 3. Static context is for what most turns need. Everything else should be discoverable when needed. Less up-front context also means less confusing or contradictory information. 4. Expect removals to win. Guardrails written for weaker models, coordination steps that became bottlenecks, and prompting for behavior the model now does on its own all cost tokens. 5. Real usage decides. Evals are a fast proxy, but they skew toward hard problems and miss the real mix of requests. ## 1. Map the harness and measure the baseline Find: - Where requests are assembled, the system prompt, and tool schemas. If a framework or SDK builds requests, find its hooks for message order, cache control, and tool loading. - How tool results are formatted, and how history is kept, trimmed, or summarized. - How subagents or parallel agents are spawned, if any. - Which models and provider APIs are used. From the provider's docs, get the prompt caching behavior (automatic or explicit breakpoints, TTL, minimum cacheable length) and the prices for output, uncached input, and cached input. - Existing logging, token accounting, and evals. If the harness doesn't record per-request token usage by billing type and cache hits, add that first. Everything later depends on it. Then render a few real requests (from logs, or by running representative tasks) and count tokens per section with the model's tokenizer or the API's usage fields. Produce: - Cost share by source × billing type. Sources: system prompt, tool definitions, skill/rule/integration descriptions, user messages, file reads, search results, command and other tool output, history, summaries, subagents. - Static tokens per request, cache hit rate, and turns per task. - Per tool: the share of runs that call it at least once, and its error rate. Read the rendered requests, not just the templates. Duplication, leaked volatile values, and misordered blocks only show up there. Rank opportunities by share of spend × fraction removable ÷ quality risk. ## 2. System prompt and injected context Label every instruction: - Keep: product or environment knowledge the model can't infer, fixes for quirks seen in this model's transcripts, and rules a mode depends on. - Rewrite: commands and emphasis into plain descriptions. Reminders into constraints: "No TODOs, no partial implementations" works better than "remember to finish implementations." Vague quantities into ranges: "generate 20–100 tasks" gets far more ambitious behavior than "generate many tasks." - Delete: things capable models do by default, guards against behavior you haven't seen from this model, text that repeats tool descriptions, and lines that could contradict a user request. Models trained to rank system instructions above user messages will side with the system prompt. - Move: anything per-user or per-request (date, environment, repo state, lists of skills or subagents, user rules) into a user-role setup message after the cache boundary. Audit other injected context the same way. As models improved, the team behind these figures dropped directory trees, pre-retrieved snippets, compressed copies of attached files, lint errors injected after every edit, forced expansion of short file reads, and caps on tool calls per turn. They kept small, high-value facts: OS, repo status, and open or recently viewed files. Skip checklists for open-ended work. The model optimizes the listed items and deprioritizes everything else. ## 3. Tool definitions Tool schemas ride along on every request. Most tools beyond the core set were each needed in under 20% of conversations, and moving them out of static context cut tool-description tokens 60%. Doing the same for integration tools (such as MCP servers), with names in context and full schemas in one folder per server that the agent can search with grep or jq, cut total tokens 46.9% in sessions that used them. - Keep in static context: high-frequency tools (for a coding agent: read, search, edit, shell), tools the model tries to call even when they're absent, and tools a mode depends on. - Offload the rest: leave a name or one-line pointer and make the full schema discoverable on demand. Group related tools so they load together, and put status (such as "needs re-authentication") where the agent will see it. - Tighten what remains: describe behavior and arguments, and drop usage lectures. - Pick the split by testing a few configurations and tracking tokens, cost, latency, tool-call errors, and task success. ## 4. Cache layout Order each request so the reusable prefix is as long as possible: `tool definitions → system instructions → [breakpoint] → setup message (skills, subagents, rules, environment) → [breakpoint] → conversation` - Keep the prefix byte-identical across turns. Use deterministic tool order and serialization, put timestamps and IDs after the boundary, and don't rewrite earlier messages except when compacting. - Use explicit breakpoints if the provider supports them. Otherwise rely on automatic prefix caching with the stable part first. Respect TTL and minimum-length rules. - Switching models mid-conversation throws away the cache (caches are per model and provider) and hands the new model a history it didn't write. When a different model is needed, run it as a subagent with fresh context. Explicit breakpoints plus moving per-request setup after them cut cold cache misses 20%. ## 5. Tool results and other context added during a run - Large outputs (commands, integrations, logs): write them to a file and return the path, size, and a short tail. The agent can tail, grep, or read ranges for more. Truncating loses data, and inlining bloats every later request. Treat long-running terminal sessions the same way. - High-volume formats: look for overhead repeated on every line or item. Numbering every 10th line of a file read instead of every line cut cache-read tokens 1.6% without hurting citation accuracy. Each number costs 3–5 tokens, and agents read tens of thousands of lines per session. Also check repeated absolute paths, verbose JSON keys, ANSI codes, progress bars, and repeated headers. - Good retrieval saves exploration turns. Adding semantic search alongside grep raised codebase question-answering accuracy 12.5% on average and cut the iterations users needed. - Tool errors waste tokens and leave confusing debris in context. Classify expected errors (invalid arguments, unexpected environment, provider error, timeout, user abort), treat unknown errors as harness bugs, and track rates per tool and per model. One focused effort along these lines cut unexpected tool errors 10×. ## 6. Long runs: compaction, subagents, and model mix - Compaction: keep the summarization prompt short and the summary compact, carry forward plan state and remaining tasks, and save the full history to a file the agent can search for details the summary dropped. A model trained to self-summarize from a one-line prompt wrote ~1k-token summaries with half the compaction error of a multi-thousand-token prompt that produced 5k+ token summaries. Untrained models may need more guidance, so test how short you can go. A more expensive summarization model made a negligible difference. - Scratchpads and running notes: rewrite them instead of appending. For repeated work in one environment, a small agent-maintained notes file with a line budget, loaded at start, is a promising way to shorten later runs. - Subagents: fresh context keeps the parent lean, but isolation adds coordination cost (duplicate or stale work). If the model already delegates on its own, remove prompting that pushes it to. Have subagents return short handoffs: what was done, findings, concerns, and deviations. A subagent should use a different model only when the user or harness says so. - Model mix: in large multi-agent runs, workers used at least 69% of tokens, and over 90% in most runs. A frontier planner with cheap workers matched a frontier model doing everything at about one-eighth the cost. Planner choice still changes worker spend. One planner that cost less on its own saw its workers use several times more tokens, and the run cost more overall. Measure the whole tree. - Routing and reasoning effort: send simple turns to a cheaper model or lower effort, and upgrade only when a stronger model is clearly better. A router built this way matched or beat single frontier models on user satisfaction at 41–68% lower cost. - Reasoning continuity: if the API returns reasoning items (including encrypted ones), pass them back on later turns and alert when they go missing. Dropping them cost one reasoning model 30% on a coding benchmark, and it burned tokens reconstructing its plan. ## 7. Fit the harness to each model Adapt to what each model was trained on instead of forcing one shape on all of them. If you've tuned the harness for a similar model, start from that version. - Edit format: use the one the model was trained on (for example, patch-style or search-and-replace). An unfamiliar format costs extra reasoning tokens and causes more mistakes. - Shell or tools: shell-first models fall back to `cat` or inline scripts. Name tools after their shell equivalents (such as `rg`), and if needed add: "If a tool exists for an action, prefer to use the tool instead of shell commands (e.g. read_file over `cat`)." - Literalness: some model families follow instructions literally and others tolerate imprecision. Some spiral on emphasized wording. Strip caps and emphasis for literal models. - Triggers: some models ignore a tool until told when to use it. A literal trigger works: "After substantive edits, use the to check recently edited files for linter errors. If you've introduced any, fix them if you can easily figure out how." - Progress updates: if a model reports progress through reasoning summaries, keep them to 1–2 sentences that note new findings or a change of tactic, and remove instructions about messaging mid-turn. - Quirks worth a targeted line: hedging or refusing as context fills ("context anxiety"), declaring completion early, stopping to ask permission, and calling tools that don't exist. Tie each added instruction to the transcript behavior it fixes. Re-audit when models change, since guidance one version needed can be dead weight for the next. ## 8. Validate - Offline: run a fixed set of realistic tasks before and after, ideally drawn from real usage and phrased the way users actually write (short and ambiguous). Compare task success, tokens, cost per task, turns, and tool errors. Don't ship a change that lowers success. - Online, if you have users: A/B test each change or small bundle. The primary metric is cost per completed task. Guardrails are task success signals, tool-call errors, latency, turns per task, and cache hit rate. For a coding agent, a good success signal is how much agent-written code survives over time. In general, check whether the user's next message moves on or reports a problem. - Ship only when cost drops and no guardrail regresses beyond noise. Record null results. ## What to change directly and what to propose - Change directly, each in its own revertible commit: token and cache telemetry, deterministic serialization and tool order, moving volatile content out of the cached prefix, explicit cache breakpoints, writing large outputs to files instead of truncating, passing back reasoning items that are being dropped, and fixes for recurring tool errors. - Change behind a flag so it can be tested: system prompt edits, tool offloading, output format changes, compaction changes, and subagent prompting. - Propose only: changes to which models run, routing, reasoning-effort defaults, or how work is split across agents. ## Traps - Asking the model to use fewer tokens or do less. - Truncating tool output. - Dropping reasoning items to save input tokens. - Volatile content in the cached prefix, or tool order that changes between requests. - Offloading a tool the model needs on the first turn or tries to call when it's missing. - Emphasis-heavy prompts (MUST, NEVER, IMPORTANT, all caps), especially with literal models. - Forcing a terser output format than the model was trained on. Fewer output tokens can mean less thinking and worse results. - Optimizing raw token counts instead of cost, per request instead of per task, or evals instead of real usage. - Switching models mid-conversation to save money. - Adding coordination layers that become bottlenecks. ## Report back with 1. The harness map and baseline: cost by source × billing type, with the biggest sources called out. 2. A ranked list of changes: layer, what changes, estimated savings and how you estimated them, quality risk, how to validate, and how to roll back. 3. The changes you made, including a system prompt diff with a keep, rewrite, delete, or move reason for each line. 4. A test plan for the flagged changes. 5. Gaps: anything you couldn't find or measure.

平时写产品技术文档，经常要做架构/流程/时序图之类 Diagram，Mermaid 默认也能做不过美观度确实一般，找了几个 Diagram Skills，分享给需要的朋友们。 1. Diagram Design 36.7K 🌟 @cathrynlavery 编辑向图解 Skill，约 39 种类型，自包含 HTML+SVG；可品牌化配色，支持从 draw. io / Mermaid / Excalidraw 重绘。做演示、文章配图优先装这个。 2. Archify 56.2K 🌟 可校验系统图 Skill；typed JSON IR → 架构/工作流/时序/数据流/生命周期交互 HTML，支持 Before/After 对比与导出。 3. Fireworks Tech Graph 11.3K 🌟 @teach_fireworks 自然语言出技术图；几何校验 SVG+PNG，可选语义 GIF；12 风格、完整 UML，内置 RAG/Multi-Agent 等模式。 4. Excalidraw Diagram Skill 4.7K 🌟 @cole_medin 生成可编辑 Excalidraw 图；强调「视觉论证」与渲染自检，适合白板协作后再手改。 5. Visual Explainer 9.7K 🌟 @nicopreme 把架构图、diff 评审、计划审计、表格等做成整页 HTML 或杂志风幻灯，偏「一页讲清楚」。