ReMAC Brings Best-of-K Exploration to Robot Control Without Entropy Tricks
ReMAC brings retry-based exploration objectives from LLM training into continuous robot control, matching SAC without any entropy bonus
20 stories
ReMAC brings retry-based exploration objectives from LLM training into continuous robot control, matching SAC without any entropy bonus
LLMCodec repurposes battle-tested video compression technology to shrink LLMs at 2-bit precision, beating existing quantization methods by 21% on downstream tasks
Researchers at King's College London test six conversational strategies for AI moral advisors, finding that helping users sit with ethical uncertainty beats both persuasion and sycophancy
AWS and Peking University researchers train 8B and 14B models that match Claude Sonnet 4.5 on literary translation by generating better training data through iterative multi-agent refinement.
Amazon FAR's LadderMan teaches a humanoid robot to climb diverse real-world ladders and perform manipulation tasks using a single motion capture reference and zero-shot sim-to-real transfer.
A new paper extends consistency training to two new internal targets and four new threat categories, showing it can unify defenses against a broader class of LLM misalignment.
A new paper proposes six entropy-based metrics that reveal how AI agents actually behave, not just whether they succeed at tasks
EEGDancer treats brain-signal emotion tracking as a game an AI agent must win, beating every existing method on three benchmarks
GenTI uses LLMs to auto-generate Snort and Suricata rules from analyst prompts, pushing zero-day attack detection from 45% to 87.4%
A new paper proves that weighted linear pooling is the only density fusion rule that stays order-invariant when executed hierarchically across distributed systems.
Researchers propose a bandit-based local harness that decouples preference learning from LLM reasoning, beating memory-augmented agents on skill selection accuracy and regret.
Researchers introduce ToolMaze, a benchmark revealing that LLM agents systematically fail to recover from tool errors — and that scaling models barely helps
A new paper identifies a tiny subset of attention heads in multimodal LLMs that do all the visual retrieval work — and shows you can use that to cut inference time in half.
HPME replaces soft edge masks with hard graph pooling and structure-level mixup, achieving up to 30% AUC gains over prior GNN explainers across 13 benchmarks.
A 30-model benchmark reveals LLMs top out at 8.6% semantic correctness on TLA+ specs, exposing a hard gap between code generation and formal verification
MARDoc replaces the bloated context window of document QA agents with a lean, structured memory that actually gets smarter across iterations
TAPO fixes a silent training flaw in GRPO that punishes good tool calls inside failed agent trajectories, delivering plug-and-play gains across three major RL algorithms
ByteDance replaced 1,000+ specialized video moderation models with a single vision-language model, cutting violation leakage by 42.7% and inference cost by 15x
A Sagnac interferometer paired with phi-OTDR and a dual-branch CNN cuts false alarms to 5% on 10-km fiber security sensing
XCount uses algebraic decision diagrams and probabilistic counting to measure how much of a tree ensemble's input space is vulnerable to sensitive feature flips — with formal guarantees.