» Tag
machine-learning
25 postsStudy: LLMs Can Transmit Hidden Traits Through Unrelated Data
Research shows LLMs can transmit behavioral traits and even misalignment to student models via data with no semantic link to that trait, like numbers.
GenRec: Netflix Moves Toward LLM-Native Recommendation Systems
Netflix's GenRec reframes recommendation as generative language modeling, signaling a shift from ranking pipelines to LLM-native architectures.
Sparse Policy Selection in RL for LLM Reasoning, Not Capability Learning
Reinforcement learning enhances LLM reasoning by focusing on sparse policy selection rather than teaching new capabilities.
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.comOpen Model Enhanced for Scientific Inquiry
The open model developed by Loka and Arcee AI enhances scientific inquiry through tool use and logical reasoning.
NVIDIA Alpamayo 2 Super: 34B Open Vision-Language-Action Model
NVIDIA introduces Alpamayo 2 Super, a 34B open vision-language-action model for robotaxis.
Knowledge Distillation: New Methods to Enhance Scalability
New methods in knowledge distillation are reducing costs and enhancing scalability for large language models.
How JAX Shards a Computation Across a Mesh
Explore how JAX improves computation placement with Auto, Explicit, and Manual methods. Learn about their impact on correctness in this detailed overview.
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.
Your AI Agent Is Gaslighting Itself: Multi-Agent Solutions May Cost 15× More
Explore the bug of self-gaslighting in AI agents and the costs of multi-agent solutions.
SparSEEty: Extracting Tokens from Sparsity-Exploiting LLM Serving Systems
SparSEEty addresses token extraction attacks on LLM serving systems, achieving high reconstruction accuracy with minimal overhead.