data/papers.json

{
  "fetched": "2026-04-17",
  "sources": [
    "arxiv-axelrod-scan-2026-04-15.json",
    "arxiv-axelrod-scan-2026-04-16.json",
    "arxiv-axelrod-scan-2026-04-17.json",
    "arxiv-calibration-scan-2026-04-15.json",
    "arxiv-calibration-scan-2026-04-16.json",
    "arxiv-calibration-scan-2026-04-17.json",
    "arxiv-multiagent-trading-scan-2026-04-15.json",
    "arxiv-multiagent-trading-scan-2026-04-16.json",
    "arxiv-multiagent-trading-scan-2026-04-17.json",
    "data/arena/proposals/ (20 iterations)",
    "data/research/*.md (11 digests)"
  ],
  "papers": [
    {
      "tag": "nomos-digest",
      "title": "Pixel World Rebuild Patterns Apr16",
      "summary": "**Date**: Apr 16, 2026 | **Analyst**: Claude Code Research | **Context**: Rebuild Nomos42 pixel-world v0.6 for clarity+understandability --- **From pixel-agents (6.7k\u2605):**",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/pixel-world-rebuild-patterns-apr16.md",
      "published": "pixel-world-rebuild-patterns-apr16",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Dashboard Redesign Apr14 2026",
      "summary": "--- --- **Reference:** bloomberg.com/professional + open-source clone github.com/feremabraz/bloomberg-terminal",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/dashboard-redesign-apr14-2026.md",
      "published": "dashboard-redesign-apr14-2026",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Dashboard Overhaul Plan Apr16",
      "summary": "**Problem:** Current dashboard shows infra + accounts + live data poorly. No single-pane-of-glass for cross-account resources (Nomos42, LBJLincoln, LBJLincoln26, TESTforge42). Live KPIs scattered. The /world iframe steals focus; /infra page is wall-of-JSON; homepage doesn't surface \"what happened today.\" **Deliverable:** Rebuild /infra, /, and /world pages using Vercel Geist + Bloomberg terminal aesthetics (per SOTA reference). Apply Moltbook Reddit-dark card grid, dYdX TradingView charting, and",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/dashboard-overhaul-plan-apr16.md",
      "published": "dashboard-overhaul-plan-apr16",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Dashboard Libraries Apr14 2026",
      "summary": "**Target:** World-class quant dashboard + 32+ agent visualization (10 traders + 22 depts) **Stack:** Next.js 15 App Router + TypeScript + @pixi/react 8 (already deployed) **Goal:** Concrete, drop-in recommendations ranked by implementation speed (week/month/rewrite)",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/dashboard-libraries-apr14-2026.md",
      "published": "dashboard-libraries-apr14-2026",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Autonomous Org Benchmark Apr14 2026",
      "summary": "Generated: 2026-04-14 | Analyst: D1-Research subagent --- **User mental model:** Paperclip-maximizer-style: propose change \u2192 measure \u2192 keep/revert, infinite loop.",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/autonomous-org-benchmark-apr14-2026.md",
      "published": "autonomous-org-benchmark-apr14-2026",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Weekly Digest 2026 04 13",
      "summary": "**Period:** 2026-04-06 to 2026-04-13 **Scans completed:** 4 days **Papers found:** 0 | **Repos found:** 14",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/weekly-digest-2026-04-13.md",
      "published": "2026-04-13",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #322 \u2014 3 proposals (68.98% to $1M)",
      "summary": "promote_model: Top model with 68% win rate; increase_aggression: Only 31.0% to $1M; restrict_categories: 2 categories with <40% win rate",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-322.json",
      "published": "2026-04-11",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Weekly Digest 2026 04 06",
      "summary": "**Period:** 2026-03-30 to 2026-04-06 **Scans completed:** 7 days **Papers found:** 0 | **Repos found:** 13",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/weekly-digest-2026-04-06.md",
      "published": "2026-04-06",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #368 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-368.json",
      "published": "2026-04-04",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #345 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-345.json",
      "published": "2026-04-04",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #319 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-319.json",
      "published": "2026-04-04",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #297 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-297.json",
      "published": "2026-04-04",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #232 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-232.json",
      "published": "2026-04-03",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #211 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-211.json",
      "published": "2026-04-03",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #185 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-185.json",
      "published": "2026-04-03",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #163 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-163.json",
      "published": "2026-04-03",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #137 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-137.json",
      "published": "2026-04-03",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #123 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-123.json",
      "published": "2026-04-03",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #74 \u2014 4 proposals (99.63% to $1M)",
      "summary": "eliminate_strategies: 2 strategies with negative ROI; promote_model: Top model with 46% win rate; increase_aggression: Only 0.4% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-74.json",
      "published": "2026-04-02",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #52 \u2014 2 proposals (69.78% to $1M)",
      "summary": "promote_model: Top model with 63% win rate; increase_aggression: Only 30.2% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-52.json",
      "published": "2026-04-02",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #44 \u2014 2 proposals (69.78% to $1M)",
      "summary": "promote_model: Top model with 63% win rate; increase_aggression: Only 30.2% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-44.json",
      "published": "2026-04-02",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #40 \u2014 2 proposals (69.78% to $1M)",
      "summary": "promote_model: Top model with 63% win rate; increase_aggression: Only 30.2% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-40.json",
      "published": "2026-04-02",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #37 \u2014 2 proposals (69.78% to $1M)",
      "summary": "promote_model: Top model with 63% win rate; increase_aggression: Only 30.2% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-37.json",
      "published": "2026-04-02",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #34 \u2014 2 proposals (69.78% to $1M)",
      "summary": "promote_model: Top model with 63% win rate; increase_aggression: Only 30.2% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-34.json",
      "published": "2026-04-02",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #25 \u2014 2 proposals (69.78% to $1M)",
      "summary": "promote_model: Top model with 63% win rate; increase_aggression: Only 30.2% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-25.json",
      "published": "2026-04-01",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-karpathy",
      "title": "Karpathy Iter #19 \u2014 2 proposals (69.78% to $1M)",
      "summary": "promote_model: Top model with 63% win rate; increase_aggression: Only 30.2% to $1M",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/arena/proposals/proposal-iter-19.json",
      "published": "2026-04-01",
      "authors": [
        "Nomos42 Karpathy Loop"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Weekly Digest 2026 03 31",
      "summary": "**Period:** 2026-03-24 to 2026-03-31 **Scans completed:** 1 days **Papers found:** 0 | **Repos found:** 10",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/weekly-digest-2026-03-31.md",
      "published": "2026-03-31",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Self Improvement Harness Sources 2026 03 31",
      "summary": "- **Authors:** Google DeepMind - **ArXiv:** [2603.03329](https://arxiv.org/abs/2603.03329) - **Date:** February 28, 2026",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/self-improvement-harness-sources-2026-03-31.md",
      "published": "2026-03-31",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Self Improvement Harness Quick Wins 2026 03 31",
      "summary": "Research on self-improving LLM harnesses (March 2026) reveals 4 **actionable, implementable techniques** to close the 0.0157 Brier gap (0.199 SOTA \u2192 0.21570 ATR). All have open-source code, peer-reviewed validation, and direct mapping to our GA/feature workflow. --- **What:** Extend `scripts/kaggle/nba_karpathy_loop.py` to only commit GA generations that beat the current best Brier.",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/self-improvement-harness-quick-wins-2026-03-31.md",
      "published": "2026-03-31",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "nomos-digest",
      "title": "Research Cycle 2026 03 26",
      "summary": "- Isotonic regression DEGRADES Brier for tree ensembles: +0.0034 worse - Replace with Beta calibration (`pip install betacal`) or Venn-Abers (`pip install venn-abers`) - **Expected: -0.003 Brier | Effort: 2h | PRIORITY: IMMEDIATE**",
      "url": "https://github.com/LBJLincoln/mon-ipad/blob/main/data/research/research-cycle-2026-03-26.md",
      "published": "2026-03-26",
      "authors": [
        "Nomos42 Research Agents"
      ],
      "source": "nomos42-internal"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Gravitational-wave lensing beyond rays: a disordered-system approach",
      "summary": "We develop a framework to describe gravitational wave propagation through a stochastic distribution of weak gravitational lenses beyond the geometric optics limit. We model the lens distribution as a static random background field and formulate the problem in the language of quenched disorder, treating the disorder averaged density matrix as the fundamental object from which observables are computed. Using the Schwinger Keldysh formalism, we construct a path-integral representation of the averag",
      "url": "http://arxiv.org/abs/2604.15313v1",
      "published": "2026-04-16T17:59:59Z",
      "authors": [
        "Ripalta Amoruso",
        "Ginevra Braga",
        "Alice Garoffolo",
        "Francescopaolo Lopez",
        "Nicola Bartolo",
        "Sabino Matarrese"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "LeapAlign: Post-Training Flow Matching Models at Any Generation Step by Building Two-Step Trajectories",
      "summary": "This paper focuses on the alignment of flow matching models with human preferences. A promising way is fine-tuning by directly backpropagating reward gradients through the differentiable generation process of flow matching. However, backpropagating through long trajectories results in prohibitive memory costs and gradient explosion. Therefore, direct-gradient methods struggle to update early generation steps, which are crucial for determining the global structure of the final image. To address t",
      "url": "http://arxiv.org/abs/2604.15311v1",
      "published": "2026-04-16T17:59:56Z",
      "authors": [
        "Zhanhao Liang",
        "Tao Yang",
        "Jie Wu",
        "Chengjian Feng",
        "Liang Zheng"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Diagnosing LLM Judge Reliability: Conformal Prediction Sets and Transitivity Violations",
      "summary": "LLM-as-judge frameworks are increasingly used for automatic NLG evaluation, yet their per-instance reliability remains poorly understood. We present a two-pronged diagnostic toolkit applied to SummEval: $\\textbf{(1)}$ a transitivity analysis that reveals widespread per-input inconsistency masked by low aggregate violation rates ($\\bar\u03c1 = 0.8$-$4.1\\%$), with $33$-$67\\%$ of documents exhibiting at least one directed 3-cycle; and $\\textbf{(2)}$ split conformal prediction sets over 1-5 Likert scores",
      "url": "http://arxiv.org/abs/2604.15302v1",
      "published": "2026-04-16T17:58:21Z",
      "authors": [
        "Manan Gupta",
        "Dhruv Kumar"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Ensembles of random quantum states tunable from volume law to area law",
      "summary": "A standard approach to generate random pure quantum states relies on sampling from the Haar measure. However, the entanglement properties of such states present a fundamental challenge for their general applicability. Here, we introduce the $\u03c3$-ensembles $\\unicode{x2013}$ a family of random quantum states with only a single control parameter. Crucially, these states are designed such that they can be tuned between volume-law and area-law behavior, which has been a major obstacle thus far. We con",
      "url": "http://arxiv.org/abs/2604.15300v1",
      "published": "2026-04-16T17:57:19Z",
      "authors": [
        "H\u00e9lo\u00efse Albot",
        "Sebastian Paeckel"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Unity and Diversity of Intracellular pH Maintenance Mechanisms",
      "summary": "All cells must sustain ionic motive forces (IMFs) -- the electrochemical gradients of permeant ions, together with the membrane potential they produce -- to regulate intracellular pH, drive secondary transport, and power ATP synthesis. Because membranes are imperfectly impermeable, IMFs continuously dissipate through passive leakage, and active transport must compensate at an energetic cost that competes with growth and biosynthesis. How environmental conditions set this cost, and why cells acro",
      "url": "http://arxiv.org/abs/2604.15296v1",
      "published": "2026-04-16T17:56:15Z",
      "authors": [
        "Guillaume Terradot",
        "Vincent Danos"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Sweet Trims are made of Threes: A c\u00e0dl\u00e0g erasure of the Brownian tree",
      "summary": "We present a simple trimming algorithm that generates nested uniform binary plane trees by removing leaves one-by-one using a best-of-three-match procedure. While its one-step transition specializes to the Luczak-Winkler & Caraceni-Stauffer coupling, its scaling limit provides a suprising c\u00e0dl\u00e0g erasure of Brownian trees, reminiscent of SLE theory.",
      "url": "http://arxiv.org/abs/2604.14138v1",
      "published": "2026-04-15T17:57:21Z",
      "authors": [
        "Alessandra Caraceni",
        "Nicolas Curien",
        "William Fleurat",
        "Adrianus Twigt"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Temporary Power Adjusting Withholding Attack",
      "summary": "We consider the block withholding attacks on pools, more specifically the state-of-the-art Power Adjusting Withholding (PAW) attack. We propose a generalization called Temporary PAW (T-PAW) where the adversary withholds a fPoW from pool mining at most $T$-time even when no other block is mined. We show that PAW attack corresponds to $T\\to\\infty$ and is not optimal. In fact, the extra reward of T-PAW compared to PAW improves by an unbounded factor as adversarial hash fraction $\u03b1$, pool size $\u03b2$ a",
      "url": "http://arxiv.org/abs/2604.14135v1",
      "published": "2026-04-15T17:55:56Z",
      "authors": [
        "Mustafa Doger",
        "Sennur Ulukus"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "AI-assisted modeling and Bayesian inference of unpolarized quark transverse momentum distributions from Drell-Yan data",
      "summary": "We present an extraction of unpolarized quark transverse-momentum-dependent parton distribution functions (TMD PDFs) from Drell-Yan data within a Bayesian inference framework, incorporating artificial intelligence at multiple stages of the analysis. Our analysis is performed at ${\\rm N^3LO}$ in perturbative QCD combined with ${\\rm N^4LL}$ resummation accuracy. We first employ an AI-driven iterative procedure to explore and rank candidate functional forms for the nonperturbative contributions to ",
      "url": "http://arxiv.org/abs/2604.14133v1",
      "published": "2026-04-15T17:54:47Z",
      "authors": [
        "Zhong-Bo Kang",
        "Luke Sellers",
        "Congyue Zhang",
        "Curtis Zhou"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "AI-assisted writing and the reorganization of scientific knowledge",
      "summary": "Generative AI systems such as ChatGPT are increasingly used in scientific writing, yet their broader implications for the organization of scientific knowledge remain unclear. We examine whether AI-assisted writing intensity, measured as the share of text in a paper that is predicted to exhibit features consistent with LLM-generated text, is associated with scientific disruption and knowledge recombination. Using approximately two million full-text research articles published between 2021 and 202",
      "url": "http://arxiv.org/abs/2604.14126v1",
      "published": "2026-04-15T17:50:16Z",
      "authors": [
        "Erjia Yan",
        "Chaoqun Ni"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Icy Volatile Enhancements in Evolving Protoplanetary Disks",
      "summary": "Protoplanetary disk ice lines shape a multitude of planet formation processes, setting the environmental composition through evolution. Ice line locations depend on molecular sublimation and deposition properties, but in dynamic disks where temperature and density structures change, so do the expected compositions of planets and planetesimals. In turbulent viscous disks with particle drift, thermal evolution, and desorption/adsorption, Price et al. 2021 demonstrated that the CO/H$_2$O ice ratio ",
      "url": "http://arxiv.org/abs/2604.14124v1",
      "published": "2026-04-15T17:49:26Z",
      "authors": [
        "Elizabeth Yunerman",
        "Ellen Price",
        "Karin \u00d6berg"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Partial majorization and Schur concave functions on the sets of quantum and classical states",
      "summary": "We construct for a Schur concave function $f$ on the set of quantum states a tight upper bound on the difference $f(\u03c1)-f(\u03c3)$ for a quantum state $\u03c1$ with finite $f(\u03c1)$ and any quantum state $\u03c3$ $m$-partially majorized by the state $\u03c1$ in the sense described in [1]. We also obtain a tight upper bound on this difference under the additional condition $\\frac{1}{2}\\|\u03c1-\u03c3\\|_1\\leq\\varepsilon$ and find simple sufficient conditions for vanishing this bound with $\\,\\min\\{\\varepsilon,1/m\\}\\to0\\,$.   The ob",
      "url": "http://arxiv.org/abs/2604.13033v1",
      "published": "2026-04-14T17:59:12Z",
      "authors": [
        "M. E. Shirokov"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Generative Refinement Networks for Visual Synthesis",
      "summary": "While diffusion models dominate the field of visual generation, they are computationally inefficient, applying a uniform computational effort regardless of different complexity. In contrast, autoregressive (AR) models are inherently complexity-aware, as evidenced by their variable likelihoods, but are often hindered by lossy discrete tokenization and error accumulation. In this work, we introduce Generative Refinement Networks (GRN), a next-generation visual synthesis paradigm to address these i",
      "url": "http://arxiv.org/abs/2604.13030v1",
      "published": "2026-04-14T17:59:03Z",
      "authors": [
        "Jian Han",
        "Jinlai Liu",
        "Jiahuan Wang",
        "Bingyue Peng",
        "Zehuan Yuan"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Conflated Inverse Modeling to Generate Diverse and Temperature-Change Inducing Urban Vegetation Patterns",
      "summary": "Urban areas are increasingly vulnerable to thermal extremes driven by rapid urbanization and climate change. Traditionally, thermal extremes have been monitored using Earth-observing satellites and numerical modeling frameworks. For example, land surface temperature derived from Landsat or Sentinel imagery is commonly used to characterize surface heating patterns. These approaches operate as forward models, translating radiative observations or modeled boundary conditions into estimates of surfa",
      "url": "http://arxiv.org/abs/2604.13028v1",
      "published": "2026-04-14T17:58:07Z",
      "authors": [
        "Baris Sarper Tezcan",
        "Hrishikesh Viswanath",
        "Rubab Saher",
        "Daniel Aliaga"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "See, Point, Refine: Multi-Turn Approach to GUI Grounding with Visual Feedback",
      "summary": "Computer Use Agents (CUAs) fundamentally rely on graphical user interface (GUI) grounding to translate language instructions into executable screen actions, but editing-level grounding in dense coding interfaces, where sub-pixel accuracy is required to interact with dense IDE elements, remains underexplored. Existing approaches typically rely on single-shot coordinate prediction, which lacks a mechanism for error correction and often fails in high-density interfaces. In this technical report, we",
      "url": "http://arxiv.org/abs/2604.13019v1",
      "published": "2026-04-14T17:55:46Z",
      "authors": [
        "Himangi Mittal",
        "Gaurav Mittal",
        "Nelson Daniel Troncoso",
        "Yu Hu"
      ],
      "source": "arxiv"
    },
    {
      "tag": "calibration-scan-2026-04",
      "title": "Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe",
      "summary": "On-policy distillation (OPD) has become a core technique in the post-training of large language models, yet its training dynamics remain poorly understood. This paper provides a systematic investigation of OPD dynamics and mechanisms. We first identify that two conditions govern whether OPD succeeds or fails: (i) the student and teacher should share compatible thinking patterns; and (ii) even with consistent thinking patterns and higher scores, the teacher must offer genuinely new capabilities b",
      "url": "http://arxiv.org/abs/2604.13016v1",
      "published": "2026-04-14T17:54:28Z",
      "authors": [
        "Yaxuan Li",
        "Yuxin Zuo",
        "Bingxiang He",
        "Jinqian Zhang",
        "Chaojun Xiao",
        "Cheng Qian",
        "Tianyu Yu",
        "Huan-ang Gao",
        "Wenkai Yang",
        "Zhiyuan Liu",
        "Ning Ding"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "When Valid Signals Fail: Regime Boundaries Between LLM Features and RL Trading Policies",
      "summary": "Can large language models (LLMs) generate continuous numerical features that improve reinforcement learning (RL) trading agents? We build a modular pipeline where a frozen LLM serves as a stateless feature extractor, transforming unstructured daily news and filings into a fixed-dimensional vector consumed by a downstream PPO agent. We introduce an automated prompt-optimization loop that treats the extraction prompt as a discrete hyperparameter and tunes it directly against the Information Coeffi",
      "url": "http://arxiv.org/abs/2604.10996v1",
      "published": "2026-04-13T04:53:06Z",
      "authors": [
        "Zhengzhe Yang"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "When Reasoning Models Hurt Behavioral Simulation: A Solver-Sampler Mismatch in Multi-Agent LLM Negotiation",
      "summary": "Large language models are increasingly used as agents in social, economic, and policy simulations. A common assumption is that stronger reasoning should improve simulation fidelity. We argue that this assumption can fail when the objective is not to solve a strategic problem, but to sample plausible boundedly rational behavior. In such settings, reasoning-enhanced models can become better solvers and worse simulators: they can over-optimize for strategically dominant actions, collapse compromise",
      "url": "http://arxiv.org/abs/2604.11840v1",
      "published": "2026-04-12T13:36:10Z",
      "authors": [
        "Sandro Andric"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "More Capable, Less Cooperative? When LLMs Fail At Zero-Cost Collaboration",
      "summary": "Large language model (LLM) agents increasingly coordinate in multi-agent systems, yet we lack an understanding of where and why cooperation failures may arise. In many real-world coordination problems, from knowledge sharing in organizations to code documentation, helping others carries negligible personal cost while generating substantial collective benefits. However, whether LLM agents cooperate when helping neither benefits nor harms the helper, while being given explicit instructions to do s",
      "url": "http://arxiv.org/abs/2604.07821v1",
      "published": "2026-04-09T05:24:27Z",
      "authors": [
        "Advait Yadav",
        "Sid Black",
        "Oliver Sourbut"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "Bounded by Risk, Not Capability: Quantifying AI Occupational Substitution Rates via a Tech-Risk Dual-Factor Model",
      "summary": "The deployment of Large Language Models (LLMs) has ignited concerns about technological unemployment. Existing task-based evaluations predominantly measure theoretical \"exposure\" to AI capabilities, ignoring critical frictions of real-world commercial adoption: liability, compliance, and physical safety. We argue occupations are not eradicated instantaneously, but gradually encroached upon via atomic actions. We introduce a Tech-Risk Dual-Factor Model to re-evaluate this. By deconstructing 923 o",
      "url": "http://arxiv.org/abs/2604.04464v1",
      "published": "2026-04-06T06:21:08Z",
      "authors": [
        "Shuyao Gao",
        "Minghao Huang"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "Single-Agent LLMs Outperform Multi-Agent Systems on Multi-Hop Reasoning Under Equal Thinking Token Budgets",
      "summary": "Recent work reports strong performance from multi-agent LLM systems (MAS), but these gains are often confounded by increased test-time computation. When computation is normalized, single-agent systems (SAS) can match or outperform MAS, yet the theoretical basis and evaluation methodology behind this comparison remain unclear. We present an information-theoretic argument, grounded in the Data Processing Inequality, suggesting that under a fixed reasoning-token budget and with perfect context util",
      "url": "http://arxiv.org/abs/2604.02460v2",
      "published": "2026-04-02T18:47:48Z",
      "authors": [
        "Dat Tran",
        "Douwe Kiela"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "A Safety-Aware Role-Orchestrated Multi-Agent LLM Framework for Behavioral Health Communication Simulation",
      "summary": "Single-agent large language model (LLM) systems struggle to simultaneously support diverse conversational functions and maintain safety in behavioral health communication. We propose a safety-aware, role-orchestrated multi-agent LLM framework designed to simulate supportive behavioral health dialogue through coordinated, role-differentiated agents. Conversational responsibilities are decomposed across specialized agents, including empathy-focused, action-oriented, and supervisory roles, while a ",
      "url": "http://arxiv.org/abs/2604.00249v1",
      "published": "2026-03-31T21:21:31Z",
      "authors": [
        "Ha Na Cho"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "Multi-Agent LLMs for Adaptive Acquisition in Bayesian Optimization",
      "summary": "The exploration-exploitation trade-off is central to sequential decision-making and black-box optimization, yet how Large Language Models (LLMs) reason about and manage this trade-off remains poorly understood. Unlike Bayesian Optimization, where exploration and exploitation are explicitly encoded through acquisition functions, LLM-based optimization relies on implicit, prompt-based reasoning over historical evaluations, making search behavior difficult to analyze or control. In this work, we pr",
      "url": "http://arxiv.org/abs/2603.28959v1",
      "published": "2026-03-30T20:05:30Z",
      "authors": [
        "Andrea Carbonati",
        "Mohammadsina Almasi",
        "Hadis Anahideh"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "Kill-Chain Canaries: Stage-Level Tracking of Prompt Injection Across Attack Surfaces and Model Safety Tiers",
      "summary": "Multi-agent LLM systems are entering production -- processing documents, managing workflows, acting on behalf of users -- yet their resilience to prompt injection is still evaluated with a single binary: did the attack succeed? This leaves architects without the diagnostic information needed to harden real pipelines. We introduce a kill-chain canary methodology that tracks a cryptographic token through four stages (EXPOSED -> PERSISTED -> RELAYED -> EXECUTED) across 950 runs, five frontier LLMs,",
      "url": "http://arxiv.org/abs/2603.28013v3",
      "published": "2026-03-30T04:07:18Z",
      "authors": [
        "Haochuan Kevin Wang",
        "Zechen Zhang"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "Prediction Arena: Benchmarking AI Models on Real-World Prediction Markets",
      "summary": "We introduce Prediction Arena, a benchmark for evaluating AI models' predictive accuracy and decision-making by enabling them to trade autonomously on live prediction markets with real capital. Unlike synthetic benchmarks, Prediction Arena tests models in environments where trades execute on actual exchanges (Kalshi and Polymarket), providing objective ground truth that cannot be gamed or overfitted. Each model operates as an independent agent starting with $10,000, making autonomous decisions e",
      "url": "http://arxiv.org/abs/2604.07355v1",
      "published": "2026-03-28T06:13:17Z",
      "authors": [
        "Jaden Zhang",
        "Gardenia Liu",
        "Oliver Johansson",
        "Hileamlak Yitayew",
        "Kamryn Ohly",
        "Grace Li"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "Benchmarking Multi-Agent LLM Architectures for Financial Document Processing: A Comparative Study of Orchestration Patterns, Cost-Accuracy Tradeoffs and Production Scaling Strategies",
      "summary": "The adoption of large language models (LLMs) for structured information extraction from financial documents has accelerated rapidly, yet production deployments face fundamental architectural decisions with limited empirical guidance. We present a systematic benchmark comparing four multi-agent orchestration architectures: sequential pipeline, parallel fan-out with merge, hierarchical supervisor-worker and reflexive self-correcting loop. These are evaluated across five frontier and open-weight LL",
      "url": "http://arxiv.org/abs/2603.22651v1",
      "published": "2026-03-24T00:02:47Z",
      "authors": [
        "Siddhant Kulkarni",
        "Yukta Kulkarni"
      ],
      "source": "arxiv"
    },
    {
      "tag": "multiagent-trading-scan-2026-04",
      "title": "TrustTrade: Human-Inspired Selective Consensus Reduces Decision Uncertainty in LLM Trading Agents",
      "summary": "Large language models (LLMs) are increasingly deployed as autonomous agents in financial trading. However, they often exhibit a hazardous behavioral bias that we term uniform trust, whereby retrieved information is implicitly assumed to be factual and heterogeneous sources are treated as equally informative. This assumption stands in sharp contrast to human decision-making, which relies on selective filtering, cross-validation, and experience-driven weighting of information sources. As a result,",
      "url": "http://arxiv.org/abs/2603.22567v1",
      "published": "2026-03-23T20:54:50Z",
      "authors": [
        "Minghan Li",
        "Rachel Gonsalves",
        "Weiyue Li",
        "Sunghoon Yoon",
        "Mengyu Wang"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "In Trust We Survive: Emergent Trust Learning",
      "summary": "We introduce Emergent Trust Learning (ETL), a lightweight, trust-based control algorithm that can be plugged into existing AI agents. It enables these to reach cooperation in competitive game environments under shared resources. Each agent maintains a compact internal trust state, which modulates memory, exploration, and action selection. ETL requires only individual rewards and local observations and incurs negligible computational and communication overhead.   We evaluate ETL in three environm",
      "url": "http://arxiv.org/abs/2603.17564v1",
      "published": "2026-03-18T10:12:54Z",
      "authors": [
        "Qianpu Chen",
        "Giulio Barbero",
        "Mike Preuss",
        "Derya Soydaner"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "The effect of a toroidal opinion space on opinion bi-polarisation",
      "summary": "Many models of opinion dynamics include measures of distance between opinions. Such models are susceptible to boundary effects where the choice of the topology of the opinion space may influence the dynamics. In this paper we study an opinion dynamics model following the seminal model by Axelrod, with the goal of understanding the effect of a toroidal opinion space. To do this we systematically compare two versions of the model: one with toroidal opinion space and one with cubic opinion space.  ",
      "url": "http://arxiv.org/abs/2603.05337v1",
      "published": "2026-03-05T16:15:45Z",
      "authors": [
        "Frank P. Pijpers",
        "Benedikt V. Meylahn",
        "Michel R. H. Mandjes"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "MO-MIX: Multi-Objective Multi-Agent Cooperative Decision-Making With Deep Reinforcement Learning",
      "summary": "Deep reinforcement learning (RL) has been applied extensively to solve complex decision-making problems. In many real-world scenarios, tasks often have several conflicting objectives and may require multiple agents to cooperate, which are the multi-objective multi-agent decision-making problems. However, only few works have been conducted on this intersection. Existing approaches are limited to separate fields and can only handle multi-agent decision-making with a single objective, or multi-obje",
      "url": "http://arxiv.org/abs/2603.00730v1",
      "published": "2026-02-28T16:25:22Z",
      "authors": [
        "Tianmeng Hu",
        "Biao Luo",
        "Chunhua Yang",
        "Tingwen Huang"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "Safe and Interpretable Multimodal Path Planning for Multi-Agent Cooperation",
      "summary": "Successful cooperation among decentralized agents requires each agent to quickly adapt its plan to the behavior of other agents. In scenarios where agents cannot confidently predict one another's intentions and plans, language communication can be crucial for ensuring safety. In this work, we focus on path-level cooperation in which agents must adapt their paths to one another in order to avoid collisions or perform physical collaboration such as joint carrying. In particular, we propose a safe ",
      "url": "http://arxiv.org/abs/2602.19304v1",
      "published": "2026-02-22T18:57:07Z",
      "authors": [
        "Haojun Shi",
        "Suyu Ye",
        "Katherine M. Guerrerio",
        "Jianzhi Shen",
        "Yifan Yin",
        "Daniel Khashabi",
        "Chien-Ming Huang",
        "Tianmin Shu"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "Characterizing Robustness of Strategies to Novelty in Zero-Sum Open Worlds",
      "summary": "In open-world environments, artificial agents must often contend with novel conditions that deviate from their training or design assumptions. This paper studies the robustness of fixed-strategy agents to such novelty within the setting of two-player zero-sum games. We present a general framework for characterizing the impact of environmental novelties, such as changes in payoff structure or action constraints, on agent performance in two distinct domains: Iterated Prisoner's Dilemma (IPD) and h",
      "url": "http://arxiv.org/abs/2602.14278v1",
      "published": "2026-02-15T19:05:04Z",
      "authors": [
        "Mayank Kejriwal",
        "Shilpa Thomas",
        "Hongyu Li"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "Altruism and Fair Objective in Mixed-Motive Markov games",
      "summary": "Cooperation is fundamental for society's viability, as it enables the emergence of structure within heterogeneous groups that seek collective well-being. However, individuals are inclined to defect in order to benefit from the group's cooperation without contributing the associated costs, thus leading to unfair situations. In game theory, social dilemmas entail this dichotomy between individual interest and collective outcome. The most dominant approach to multi-agent cooperation is the utilitar",
      "url": "http://arxiv.org/abs/2602.08389v1",
      "published": "2026-02-09T08:40:52Z",
      "authors": [
        "Yao-hua Franck Xu",
        "Tayeb Lemlouma",
        "Arnaud Braud",
        "Jean-Marie Bonnin"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "Emergent Cooperation in Quantum Multi-Agent Reinforcement Learning Using Communication",
      "summary": "Emergent cooperation in classical Multi-Agent Reinforcement Learning has gained significant attention, particularly in the context of Sequential Social Dilemmas (SSDs). While classical reinforcement learning approaches have demonstrated capability for emergent cooperation, research on extending these methods to Quantum Multi-Agent Reinforcement Learning remains limited, particularly through communication. In this paper, we apply communication approaches to quantum Q-Learning agents: the Mutual A",
      "url": "http://arxiv.org/abs/2601.18419v1",
      "published": "2026-01-26T12:21:05Z",
      "authors": [
        "Michael K\u00f6lle",
        "Christian Reff",
        "Leo S\u00fcnkel",
        "Julian Hager",
        "Gerhard Stenzel",
        "Claudia Linnhoff-Popien"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "Multi-Agent Cooperative Learning for Robust Vision-Language Alignment under OOD Concepts",
      "summary": "This paper introduces a novel Multi-Agent Cooperative Learning (MACL) framework to address cross-modal alignment collapse in vision-language models when handling out-of-distribution (OOD) concepts. Four core agents, including image, text, name, and coordination agents, collaboratively mitigate modality imbalance through structured message passing. The proposed framework enables multi-agent feature space name learning, incorporates a context exchange enhanced few-shot learning algorithm, and adop",
      "url": "http://arxiv.org/abs/2601.09746v1",
      "published": "2026-01-11T20:36:47Z",
      "authors": [
        "Philip Xu"
      ],
      "source": "arxiv"
    },
    {
      "tag": "axelrod-scan-2026-04",
      "title": "Evolving Personalities in Chaos: An LLM-Augmented Framework for Character Discovery in the Iterated Prisoners Dilemma under Environmental Stress",
      "summary": "Standard simulations of the Iterated Prisoners Dilemma (IPD) operate in deterministic, noise-free environments, producing strategies that may be theoretically optimal but fragile when confronted with real-world uncertainty. This paper addresses two critical gaps in evolutionary game theory research: (1) the absence of realistic environmental stressors during strategy evolution, and (2) the Interpretability Gap, where evolved genetic strategies remain opaque binary sequences devoid of semantic me",
      "url": "http://arxiv.org/abs/2601.02407v1",
      "published": "2026-01-01T18:34:05Z",
      "authors": [
        "Oguzhan Yildirim"
      ],
      "source": "arxiv"
    }
  ]
}