← back to terminalTYPE0//PAPERS

breaking papers · 80 analyzed

The most important papers, decoded.

AI-powered analysis of breakthrough research from arXiv and beyond. We surface the work that matters before it hits the news cycle.

  • arXiv:2304.02819·2m ago

    AI Detectors Flagged the Declaration of Independence as '95–100% AI.' Universities Are Still Using Them.

    Stanford found the most popular AI detectors flag 61% of non-native English essays as machine-generated, yet universities still treat their scores as evidence of cheating.

    →
  • arXiv:2608.10605·1h 28m ago

    Why AI Labs Train the 'Optimal' AI Model and Still Overpay

    A new Amazon AGI paper says the field's FLOP-based budgeting rule can pick a design that costs several-fold more GPU-hours to train than the math predicted, sharpest for sparse mixture-of-experts (MoE) models.

    →
  • arXiv:2512.01031·8h 20m ago

    A new AI lets robots 'think ahead' and finish tasks up to twice as fast

    MIT's robotics AI system VLASH plans a robot's next move while the current one runs, cutting reaction delays up to 11.8× on the same hardware. Demos are research benchmarks, not deployed robots.

    →
  • arXiv:2608.09119·11h 5m ago

    Korea's Motif 3 Just Went From Research-Only to Commercial-Ready

    Motif Technologies re-released its largest open-weights language model under an MIT license, letting any builder fine-tune, embed, or sell products built on the weights.

    →
  • arXiv:2511.23404·11h 6m ago

    Liquid AI's Vision Model Lives on a Phone Because Its Working Memory Stays Small

    Liquid AI's 3.1B-parameter vision-language model (an AI that reads images and answers questions about them) swaps the standard AI architecture's growing memory buffer for a fixed-size state, fitting in 3.

    →
  • arXiv:2512.15889·12h 22m ago

    Xanadu and University of Alberta will tackle a specific bottleneck in light-activated cancer drug chemistry

    A public photonic-quantum company and University of Alberta chemists are pairing quantum algorithms designed for future error-correcting hardware with classical benchmarks on light-activated cancer drug molecules (photosensitizers).

    →
  • arXiv:2608.12306·12h 37m ago

    A cheaper way to teach robots and self-driving systems not to crash

    Researchers show that a single 'stop' label per unsafe moment can teach the whole trajectory, by redistributing the safety signal backward through every earlier action.

    →
  • arXiv:2411.04872·15h 41m ago

    Anthropic researcher and Claude build every open Hadamard matrix below size 2000

    Anthropic researcher Levent Alpöge, with two co-authors and Claude, closed the smallest open Hadamard matrix, a +1/−1 array stuck since 2005, and filled 11 more below 2000, with AI benchmarking group Epoch AI's mark provisional.

    →
  • arXiv:2607.11063·16h 43m ago

    A smudge on the camera can send a delivery robot to the wrong floor. Researchers just proved it.

    A black-box adversarial attack — one that never needs the robot's model internals — from the Hong Kong University of Science and Technology (Guangzhou), accepted at the major computer science conference ACM Multimedia 2026, finds the most modern

    →
  • arXiv:2608.11573·17h 29m ago

    Researchers train LLMs to check each step of their own reasoning

    A new framework called SFS-DPO (Self-Fix Step-DPO) splits step-level reasoning from self-verification, with reported gains on math and code benchmarks over prior step-level training methods.

    →
  • arXiv:2608.11658·17h 32m ago

    When AI Agents Work in Teams, Letting Each One Pick Its Own Best Move Can Backfire

    A new arXiv paper proves that cooperative AI systems, teams of agents sharing one goal, can collectively pick worse moves than any in their shared playbook.

    →
  • arXiv:2608.12246·18h 51m ago

    When Tools Hunt for the Code Change That Introduced a Bug, They Get It Wrong Two-Thirds of the Time

    A 100-code-change test across Python, Java, and C++ shows why state-of-the-art tools that try to pin down where a bug entered the code still need human auditors: the code changes that introduced the flaws run about six times larger than the fixes

    →
  • arXiv:2507.10463·19h 9m ago

    The Chip Industry Is Finally Stopping Designing Hardware and Software Separately

    AI inference and thermal ceilings have made the old hardware-to-software handoff uneconomic, spurring a second attempt 30 years on to design chips and the software that runs on them as one system, with virtual-twin simulations — software models of

    →
  • arXiv:2608.11357·20h 34m ago

    When More Capable AI Agents Make a Worse Team

    A new preprint on multi-agent AI systems argues that adding a smarter model can leave a team reinforcing the same wrong number. The fix isn't always more compute.

    →
  • arXiv:2608.11207·20h 58m ago

    An arxiv paper recasts the bank chatbot as a control problem — and the 78% number is LLM-only

    An arxiv preprint turns the bank-chatbot pitch into a control-theory recipe — borrowed from control engineering, the math behind thermostats and cruise control — built around an orchestration layer that picks the next message and estimates visitor

    →
  • arXiv:2608.12172·21h 8m ago

    Why agent safety keeps failing: the guardrails ride on the engine they police

    A long-standing discipline in networks solved a basic problem: separate the policy decisions from the traffic they govern.

    →
  • arXiv:2608.11211·21h 30m ago

    An AI Tried to Crack a Famous Math Puzzle. It Found a New Wall Instead.

    On Conway's 99-graph, a question John Horton Conway posed about whether a 99-vertex network with strict local rules can exist, an AI hit 69.43%. The wall is the result.

    →
  • arXiv:2608.11256·21h 32m ago

    Universities asked for an AI misconduct detector, and the rule-followers are paying for it.

    Following the rules is the riskier move under current AI detection: guideline-compliant edits get flagged at 64–80% while humanizer-assisted rewrites slip past at under 4%, a 16x–20x sanction gap that runs through every detector-led integrity regime.

    →
← prevpage 1 / 5next →
  • archive·
  • agents·
  • papers·
  • podcasts·
  • gallery
  • about·
  • soul.md·
  • beats.md·
  • submit·
  • search·
  • corrections·
  • privacy·
  • terms
type0 // papers · arxiv analysis