» Tag
llm
48 postsMCP Security: Trust Boundaries and the Attacks the Spec Won't Stop
A breakdown of MCP's four trust boundaries, stdio server risks, OAuth 2.1/PKCE/CIMD updates, and the confused-deputy attack class engineers must handle.
TutorMoments: Testing If AI Tutors Know When to Help or Hold Back
Allen AI's TutorMoments benchmark tests whether LLM tutors know when to scaffold and when to push students toward harder reasoning.
SIMURG: Real-Time Guard Stops LLM Decoding Corruption Mid-Stream
SIMURG is a numpy-only real-time monitor that detects and aborts LLM decoding corruption mid-stream, with zero training and conformal calibration.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comRestore LLM Inference Capacity Quickly with Shadow Engine Recovery in NVIDIA Dynamo
NVIDIA Dynamo's shadow engine recovery allows rapid restoration of LLM processes after failures.
Cross-vendor byte-identical inference for a 72B LLM (AMD MI300X vs. Nvidia H100)
A new protocol enables byte-identical outputs from a 72B LLM using AMD MI300X and Nvidia H100.
Stealing Reasoning Traces from Proprietary LLM APIs
We found a way to extract reasoning traces from frontier AI APIs, highlighting significant security risks and potential data leaks for engineers.
An Open Agent Security Benchmark: Uncaught Attacks
An open benchmark featuring 497 attacks targeting modern LLM agents has been established.
Context Engineering in LLM Applications
Context engineering optimizes LLMs with domain knowledge for better performance. It uses an index for more accurate responses.
LLM Debugger: Identifying Fine-Tuning Failures
Gradian is a training data attribution engine for LoRA fine-tuned LLMs, simplifying the process of identifying and correcting errors.
Meta Doubles Advertising Efficiency with GEM Training Innovations
Meta has doubled training efficiency for its GEM model in ad recommendations, overcoming engineering challenges through innovative solutions.