federated probe-logit distillation rate limits under heterogeneous bandwidth
new lower bounds show that the bandwidth term in federated probe-logit distillation is tight, and the method extends to nodes with different upload budgets.
author
baris
Baris operates SummarizedData and maintains its sources, publishing rules, and automation. Read the editorial process.
ai agent plans molecular tweaks for drug lead optimization
a new llm-based agent called trace uses tool planning to optimize drug-like molecules over multiple steps, improving properties while keeping key structures intact.
behavior-aware corrections stabilize off-policy td learning
replacing the auxiliary covariance matrix with the behavior bellman matrix improves stability in off-policy temporal-difference learning.
anytime-valid federated conformal rag for llm swarms
a new method extends federated conformal rag to provide valid uncertainty estimates at any stopping time, enabling safer adaptive control in distributed language model systems.
claude opus 4.8 ships with honesty improvements
anthropic releases claude opus 4.8, a minor update focusing on reduced hallucinations and mid-conversation system messages.
frontier models score below 50% on enterprise it agent benchmark
itbench-aa evaluates ai agents on kubernetes incident response, with claude opus 4.7 leading at 47% accuracy.
google research at i/o 2026: ai for science, health, and more
google research highlights from i/o 2026 include new ai tools for scientific discovery, health coaching, and edge computing, plus advances in weather prediction and model factuality.
seven ai projects to automate real workflows in 2026
build practical ai assistants for job search, research, invoice processing, and more with step-by-step guides.
aws rebuilds search for ai agents
aws launches opensearch serverless to handle spiky agent traffic, scaling compute from zero to meet bursts and cutting idle costs.
ai token futures trading is coming
exchanges are developing derivatives markets for llm tokens, letting businesses hedge against compute costs.
asana buys no-code agent builder stack ai
asana acquires stack ai to integrate no-code ai agents into its work management platform, aiming to automate complex business processes.
anthropic raises $65b at $965b valuation before ipo
anthropic closed a $65 billion series h round, reaching a $965 billion valuation as it prepares to go public.
tuning local llm settings with ollama
learn to fine-tune local language model parameters, optimize hardware, and format prompts using ollama's modelfile, environment variables, and go templates.
sesame launches ios app for conversational ai agents
sesame, founded by oculus creators, releases an ios app with ai agents that talk naturally, pause to think, and adapt mid-sentence.
dynaschedbench calibrates dynamic scheduling benchmarks for llm agents
a new framework controls instance difficulty to test llm-based scheduling agents, revealing an observability paradox where more information can hurt performance.
rsi is the new agi — and just as hard to pin down
recursive self-improvement has become the latest buzzword in ai, with startups and researchers chasing systems that can upgrade themselves without human help.
leaked renders show apple's new siri app and ai plans
bloomberg reveals apple's upcoming siri app and ai integration in ios 27, aiming to rival chatgpt with google gemini-powered search and a standalone chatbot app.
google io 2026: 12 big ai reveals
google io 2026 unveiled gemini omni for video generation, search agents, and new ai tools across products.