Search: “context”

retrieval is a decision problem, not just a similarity problem cool experiment from @assaf_elovic swapping embeddings for Jev in GPT Researcher - 73% vs 46% relevant context decision models will show

🛠️ MiMo-V2.6 update: tool-call repetition, diagnosed & fixed. After the MiMo-V2.6 series models launched, we noticed them sometimes repeating identical or highly similar tool calls — burning context a

Cool paper showing how effective tuning a system prompt for an agent can be. Recommended paper if you tune agent harnesses. This paper presents EvolveTrade, which treats a trading agent's system promp

Interesting work on long-horizon research agents. PrimeScientists can decide where a research agent spends its budget. Achieves 10.3% more reward with 50.6% fewer research attempts, under the same bud

Meet Qwen3.8-LiveTranslate, Qwen's next-generation real-time simultaneous interpretation model! 📢 Built on an Interleave architecture, it improves faithfulness, fluency, and conciseness while reducing

Agentic-SDLC requires continuous learning 🎙️ New episode of The Agentic Review is live. @itamar_mar and @nnennahacks sit down with Emilie Schario, VP of Engineering at @anacondainc, to dive into AI RO

Every answer in Perplexity starts with embedding and ranking models picking the most relevant results for the query. Today we published research on how we built SoTA serving infrastructure behind thos

Brilliant paper on long-horizon agents. They cut 78.9% of an agent's LLM calls while raising its success rate. Here is how: It turns out that ReAct issues one primitive action per model round. That al

Excited to welcome @greptile, the AI code reviewer trusted by 11,000+ teams, as a Together AI customer. Greptile's agents review and test pull requests with full context of your codebase, and running

Great paper from Google and colleagues. It proposes an interesting approach to improve agents on long-horizon tasks. Long-running agents slow down and start poisoning their own context. Both symptoms

LlamaParse is officially a verified connector on Claude! 🦙 Claude is perfectly fine for the occasional text-only PDF. But most of the context your agent needs lives in messy PDFs, spreadsheets, scanne

The Top AI Papers of the Week (August 24 - 30): - Skill Lift - JIT-Agent - Prime Agent - Judges as a Lifecycle - Co-Scientist in Real Labs - What Compaction Destroys - Context Management as Code Read

Impressive work from Alibaba. (bookmark it) If you build long-running agents and keep rewriting your memory schema, take a look at this approach. It basically treats agent context management as a prog

Meta returns to open weights: Muse Glimmer, its first open-weights release since Llama 4, scores 35 on the Artificial Analysis Intelligence Index. It is a 30B-parameter model, and the first from Meta

Finally a good paper testing whether agent memory needs an LLM at all. Production memory stacks spend extra model calls on summarizing interactions, writing records, and reranking retrievals. Every on

OpenWiki now connects to LangSmith traces to analyze how coding agents interact with your repo during wiki generations! We added a LangSmith tracing connector so OpenWiki can retrieve more context int

Health in ChatGPT is starting to roll out to U.S. users. You can securely connect Apple Health and supported medical records to understand your information in context, track what has changed, and have

nvidia ai just handed dgx spark owners a real gift. nemotron labs 3 puzzle 75b a9b nvfp4 is basically built for this box. 75b total, 9.3b active, nvfp4, mamba plus moe, 256k context in the config, and

Striking paper from Wharton. The big conclusion: AI must increase productivity 2.7x -- and quickly -- or tech companies risk bankruptcy with all that entails for the economy. For context: this is how

Startup discovery platform @harmonic_ai rebuilt Scout, their AI platform using Deep Agents and LangSmith. Deep Agents: One frontier model + two tool sets (global company data and firm-specific context

I'm proud to share that @Glean has surpassed 300M ARR, just five months after crossing 200M and growing ~3x over the past 15 months. This is an exciting milestone for Glean, and it's a signal about wh

The MiniMax M2 series was one of the most widely used open-weight LLM series earlier this year. Now, we got a technical report with some interesting tidbits. I summarized some of them below: 1. Full a

This post from Gary Marcus appears to reference a humorous or notable moment related to his work, with image attachments in Traditional Chinese that provide context to the claim that he 'has won again

I cannot provide an accurate summary of this entry as the title appears to be truncated mid-sentence and lacks sufficient context. Based on the incomplete title, it likely discusses how software speci

💡 Why did @togethercompute choose NVIDIA Blackwell to serve DeepSeek-V4? Because NVIDIA Blackwell is built for the bottlenecks that matter most in long-context inference: → KV-cache pressure during de

Swyx posted a tweet referencing Vivian Bala with a statement about demonstrating something to the world, though the specific context of what will be demonstrated is not clear from the title alone. The

Strong Opinions, Loosely Held on Agent + Harness Engineering: 1. You can outperform any default harness+model (including codex & claude code) on pretty much any Task by engineering the harness around

communication is hard! structured outputs make its more clear how subagents should communicate back to the main agent deepagents now supports structured output for subagents! an under appreciated piec

@hwchase17 is right, memory creates lock-in. But not just conversations. Tool registries, hooks, agent conventions -> that's harness memory too. In my setup I define agent tools once in a shared regis

memory is core to how you get enhanced output. it is key to defensibility across AI products: 1. increase personalization and therefore lock-in 2. benefit from distribution through multiplayer-AI inte

This post is so so good. We are at this interesting point where we’ve started to really figure out ‘memory’ with these LLM-based systems. And @hwchase17 is totally right - memory is basically just con

New course: Efficient Inference with SGLang: Text and Image Generation, built in partnership with LMSys @lmsysorg and RadixArk @radixark, and taught by Richard Chen @richardczl, a Member of Technical

The memory ownership point is the real story here. Running a persistent agent 24/7, memory isn't a feature — it's the entire value layer. Lose it and you're back to day zero. Open harness + portable m

The specific post at status ID `2041927488918413589` could not be retrieved directly, as it did not appear in search results. However, based on the surrounding context from the same account (@Vtriv...

router lives in the harness, not a gateway, bc the harness knows the task blog: https://www.langchain.com/blog/how-to-build-a-model-router-in-the-harness https://x.com/neil_xbt/status/2109979966859510

Big new paper nobody is talking about ‼️ 🌎 Robot world models now have scaling laws. Meta trained an 8B-parameter JEPA on 15,000 hours of robot video to prove it. RoboJEPA (FAIR at Meta + Mila) imagin

Agentic OCR refers to an autonomous OCR framework that treats text recognition as a dynamic, context‑aware process rather than a static extraction step. The article discusses lessons learned from buil

PDFs aren't built for machines: text shows up as glyphs with coordinates, tables as line segments instead of table structures. @jerryjliu0, CEO of LlamaIndex, digs into what that means for AI agents i

Been integrating Jev into my custom harness. I couldn't be more excited about the results I am seeing. Most demos on my timeline are flashy but very basic. Notice they are all about some boring classi

Super interesting work from Zoom and colleagues. If you maintain a hand-built coding harness, there are some great insights here. (bookmark it) They held the execution loop of a coding harness fixed a

If you build agent harnesses, this is important. Should you avoid subagents, or can they be useful? My thoughts as a harness builder: I remember using subagents in Claude Code, and I mostly found them

deepseek v4.1 flash has arrived on together ai it beats gpt-5.6 sol on agentic benchmarks at one third the cost per task the model supports a 1m context window and native multimodal input, with 552b t

// human-values.pseudo | ARTIFACT 0.1.0-experimental.1 | BASE 1.2.0-rc1 | EDITION ai-compact | NARRATOR System_Synthesis_Node_01 | PROVENANCE human-directed, AI-assisted | STATUS single-encoder resear

Recommended read. Interaction horizon scheduling is an underexplored control problem in agentic RL This paper from the Qwen team takes a closer look at the problem. Scaling the maximum number of envir

Good work on improving memory for long-horizon agents. They separate two things that agent memory papers usually collapse into one. How memories get merged when they are written, and how retrieved con

correction: the switching costs weren't enough to stop my desire to try out gpt-6 astra i ported over all my skills and conversational context to codex this weekend. planning to use it as my daily dri