Looksmaxxing2h ago4 min read
RoboFake Paper Tests Deepfake Detectors on Robot Faces
An arXiv study finds many deepfake detectors fail when synthetic faces appear on humanoid robot displays.
Cyber Security2h ago4 min read
BARBICAN Gateway Paper Splits MCP Prompt Injection Fixes
A layered gateway preprint labels attack segments so agents can delete, swap, or repair malicious tool context.
Supply Chain3h ago4 min read
RAGLeak Study Flags Supply Chain Backdoors in Enterprise RAG
A security preprint shows poisoned fine tuning can leak private retrieval data while bypassing common defenses.
Agri Tech4h ago4 min read
Agri SCHOLAR Paper Grounds Crop Advice in Simulation
An IISc preprint pairs crop simulators with bounded LLM optimization for rolling in season farm guidance.
Latest this week
Blockchain4h ago4 min read
PQ Vault Preprint Sketches Quantum Safe Ethereum Custody
Researchers propose a vault contract that mixes hash checks and post quantum signatures for long horizon custody.
Biotech5h ago4 min read
ProtGlycanDock Preprint Targets Protein Sugar Docking Bench
A new arXiv paper offers a dedicated protein glycan docking model and benchmark for biologics teams.
Robotics6h ago4 min read
RoboJEPA Study Links Robot World Model Scaling to Real Plans
A Meta FAIR led preprint reports scaling laws for an eight billion parameter robot world model with public checkpoints.
Quantum6h ago4 min read
CoDecoder Paper Targets Faster Quantum Error Decoding
A Colorado Boulder preprint reports a neural front end that can speed some quantum decoders up to about 100 times.
AI7h ago4 min read
Agent Benchmark Paper Says Setup Choices Beat Bigger Models
A Munich and Inria preprint shows agent configuration often beats model size on scientific coding tasks.
LooksmaxxingOct 26, 20264 min read
PAL Net three D landmarks power facial symmetry test pipelines
PAL Net in Informatics in Medicine Unlocked improves three D facial landmark detection for face analysis AI and facial symmetry test apps built on scanned meshes.
Cyber SecurityOct 26, 20264 min read
AgentBound paper pushes permission manifests for MCP agent tools
Salvaneschi team arXiv work AgentBound reports about eighty point nine percent accurate MCP policy manifests across two hundred ninety six servers with lightweight enforcement.
Supply ChainOct 26, 20264 min read
MoSES mixes LoRA routing experts for sixteen vehicle routing variants
NeurIPS 2025 paper MoSES on arXiv uses gated LoRA experts so one neural solver tackles sixteen vehicle routing variants with open PyTorch code.
AIOct 25, 20264 min read
LeCun Team H JEPA Study Raises Maze Success From 18 to 73 Percent
A new arXiv preprint stacks world models in layers for long tasks. Turing laureate Yann LeCun co authored the October 2026 work.
AIOct 22, 20262 min read
Sakana Continuous Memory Machine Splits Short and Long AI Memory
Sakana AI posted a two bank memory design on arXiv that may help agents keep fresh notes and older facts without one crowded loop.
AIOct 21, 20264 min read
Context Language Models Let Agents Edit Their Own Memory File
Meta FAIR and partners posted a preprint on models that trim their own context file during long tasks, reporting higher accuracy and lower compute on two benchmarks.
QuantumOct 25, 20264 min read
OpenAI Math Preprint Shows Quantum Search Can Beat Classical By Nearly Fourth Power
An October 2026 theory note builds functions with a wide quantum speedup. It caps a long debate on query complexity exponents.
QuantumOct 22, 20264 min read
Local Measurement Theory Clarifies Readouts on Quantum Cellular Automata
A 19 October 2026 arXiv preprint spells out what local measurements can reveal on quantum cellular automata grids.
QuantumOct 21, 20264 min read
Ultra High Rate Quantum Codes Pack More Logical Qubits Per Block
An arXiv preprint gives design rules and parameter tables for compact quantum error codes with rates near 0.21 to 0.23 at useful distances.
RoboticsOct 25, 20264 min read
HiRe MoMa Adds Human Nudges To Boost TIAGo Robot Success On Messy Tasks
An October 2026 arXiv paper adds small motion fixes on top of a frozen policy. TIAGo place success reached 95 percent after short tuning.
RoboticsOct 22, 20264 min read
EmbodiedSmith Loop Raises Sim Task Success on Robot Demo Benchmark
EmbodiedSmith co evolves sim scenes and robot tasks and reports higher success on hard jobs in an October 2026 preprint.
RoboticsOct 21, 20264 min read
Industrial ReMEmbR Shares Spatiotemporal Memory Across Robot Fleets
NVIDIA linked researchers posted a preprint on fleet scale memory with captions, timestamps, and poses for cross robot warehouse style queries.
BiotechOct 25, 20264 min read
NuCaliby Designs DNA Sequences That Still Fold Into Target Proteins
A 5 October 2026 bioRxiv preprint designs coding DNA for a target protein shape. Authors report NeurIPS 2026 acceptance.
BiotechOct 22, 20264 min read
Blitz Preprint Targets Faster Diffusion Steps in AlphaFold3 and Boltz 2
Blitz aims to speed the denoising core shared by AlphaFold3 and Boltz 2 structure pipelines in a 16 October 2026 preprint.
BiotechOct 21, 20264 min read
PREpiBind Uses Protein Language Models for MHC Class II Binding
A Seoul team bioRxiv preprint pairs peptide and HLA embeddings to rank peptide binders, with public code and weights for lab triage.
BlockchainOct 25, 20262 min read
Ci Poseidon Hash Aims For Cleaner Constants In Zero Knowledge Proofs
An IACR ePrint proposes a ZK friendly hash with simple rational constants. Readme tables report near fifty percent avalanche on major fields.
BlockchainOct 22, 20264 min read
Behavior Aware Fraud Checks Proposed for AI Agent Payment Flows
An October 2026 preprint pairs behavior aware fraud scoring with controls on AI agent payment authorization.
BlockchainOct 21, 20264 min read
Gryphes Hybrid Proofs Split zkRollup Logic and Hash Heavy Checks
HKUST and OKX researchers posted an IACR ePrint on modular SNARKs that mix matrix lookup with hash tuned proofs for rollups.
Agri TechOct 25, 20264 min read
DLCAF LLM Pipeline Turns Tomato Leaf Photos Into Greenhouse Action Plans
A 2026 journal study tags tomato leaf disease and maps rows over time. An LLM module drafts short greenhouse advisories from those maps.
Agri TechOct 22, 20264 min read
GRAFT Model Answers Tomato Leaf Disease Questions From Photos
GRAFT uses visual question answering so growers can ask plain language questions about tomato leaf disease from photos.
Agri TechOct 21, 20264 min read
AgroGround Benchmark Makes Farm AI Point to Lesions on Leaves
An arXiv preprint merges eight ag vision sources into nearly 795k grounded tasks so models must name diseases and show evidence on the plant.
Supply ChainOct 22, 20264 min read
Columbia Team Studies Blocking Model Theft From Planning APIs
Researchers study defenses against model extraction from contextual linear optimization APIs used in routing and sourcing math.
Supply ChainOct 21, 20264 min read
Neural Operator Planning Under Noise Targets Uncertain Supply Decisions
An arXiv preprint applies neural operators to planning when transitions are noisy, a pattern inventory and logistics teams simulate often.
Cyber SecurityOct 22, 20264 min read
Long Term Agent Memory Can Extend Prompt Injection Harm Study Warns
A 20 October 2026 preprint warns that long term agent memory can turn one prompt injection into lasting compromise.
Cyber SecurityOct 21, 20264 min read
Distributed Multi Agent Attacks Hide Harmful Goals From Monitors
Security researchers reported a frontier monitor caught about 7 percent of split agent attacks versus 87 percent when one agent did the whole job.
LooksmaxxingOct 22, 20264 min read
Face Analysis AI Gains Edge Friendly 3D Gaussian Head Avatar Research
A 7 October 2026 preprint explores efficient 3D Gaussian head avatars for face analysis AI on phones and edge devices.
LooksmaxxingOct 21, 20264 min read
Face Analysis AI Study Matches Photos to Device Measured Skin Values
Peer reviewed skin research on 1,099 Korean adults shows how face analysis AI tracks visual grades and lab device readings across three camera types.
