NumLeak: Public Numeric Benchmarks as Latent Labels in Foundation Models

• URL: http://arxiv.org/abs/2605.30393v1
• Date: Thu, 28 May 2026 12:52:49 GMT
• Title: NumLeak: Public Numeric Benchmarks as Latent Labels in Foundation Models
• Abstract summary: NumLeak is a measurement framework that combines API-boundary probes on production models with a white-box controlled validation on an open causal LM.<n>Top-tier frontier LLMs recall the Fama-French market excess return at 3-seed pooled Pearson r=0.97-0.99 while staying within 0.15 within-25bps on the five sibling factors.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Public numeric benchmarks appear in pretraining, so an evaluation that conditions on a date may be measuring memorized recall rather than out-of-sample skill. We introduce NumLeak, a measurement framework that combines API-boundary probes on production models with a white-box controlled validation on an open causal LM. Top-tier frontier LLMs recall the Fama-French market excess return at 3-seed pooled Pearson r=0.97-0.99 while staying within 0.15 within-25bps on the five sibling factors; comparable fidelity appears on U.S. unemployment, CPI inflation, and NOAA temperature. On a recent-release holdout, parse rate collapses to 21-57% but r stays at approximately 0.99 on months answered, the refuse-or-recall asymmetry a memorized channel predicts. The white-box experiment reproduces the dose-response, and logprob ranking detects memorization that open-ended generation misses, implying closed-API black-box probes understate the channel. A Sonnet "date to market-sentiment" regression that correlates with true Mkt-RF at r=0.74 collapses to r=0.02 once the model's own recall is residualized out. A one-line system-prompt defense blocks 99.8% of a non-adaptive single-turn suffix attack set at near-zero utility cost on conceptual and historical-narrative queries

Related papers

• Agent Behavioral Contracts II: Certifying Compositional Reliability Without Assuming Independence [0.0]
  Two instances of one model, in a two-agent handoff, co-fail on 90.0% of the missions on which either fails.<n>Substituting a different model reduces the association in six of six contrasts.
  arXiv  Detail & Related papers  (2026-08-13T07:25:05Z)
• Accurate Ensembles, Fragile Narratives: Multi-Scale Stacking and a Fidelity Audit of LLM-Generated Explanations for Credit Risk [0.0]
  Credit scoring increasingly relies on models whose decision logic cannot be read off their parameters.<n>A common proposal closes that gap with a language model: compute feature attributions, hand them to an LLM, and let it write the rationale.<n>We build such a system end to end and test whether the second half of the promise holds.
  arXiv  Detail & Related papers  (2026-08-08T13:22:14Z)
• The Low Frequency Trap: Video Language Models Fail at Simple Event Bookkeeping [70.07961857926377]
  We introduce trace-grounded parametric profiling for event counting in three controlled video tasks.<n>Across 2,190 videos, we vary event count N and frequency F while holding rendering fixed.<n>At an 80% reliability threshold, Gemini 3.6 Flash reliably counts persistent state transitions up to 12 events at 0.5 and 1.0 Hz, yet demonstrates no reliable positive-count region for transient blinking events.
  arXiv  Detail & Related papers  (2026-08-06T17:57:06Z)
• Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion [46.262619407930266]
  We train a multi-label classifier on a curated gold dataset of 1,207 CVEs from expert MITRE Center for Threat-Informed Defense mappings.<n>The resulting model approximately doubles recall@5 compared with a zero-shot embedding-similarity baseline.<n>We then investigate whether LLM-assisted labeling can extend the gold dataset.
  arXiv  Detail & Related papers  (2026-07-28T10:59:04Z)
• Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade [14.63102636181275]
  Large language model (LLM) agents often waste inference compute by continuing multi-step trajectories that are already doomed to fail.<n>We study early failure prediction and inference-time early stopping for LLM agents using hidden-state probes.<n>We turn this signal into a recall-controlled abort cascade for reducing LLM agent inference costs.
  arXiv  Detail & Related papers  (2026-07-07T17:03:55Z)
• AI Trading's Alpha Singularity: Emergent Market Reasoning through Agent-to-Agent Self-Evolution [11.790070440436294]
  Alpha mining holds the scoring function fixed and varies the search algorithm over it.<n>A search that converges against a fixed scorer overfits whatever the scorer cannot penalize, a primary cause of the out-of-sample generalization gap.<n>We treat the scoring function as a search artifact alongside the alpha factors and study what conditions make this joint search admissible.
  arXiv  Detail & Related papers  (2026-06-28T04:41:00Z)
• Catching One in Five: LLM-as-Judge Blind Spots in Production Multi-Turn Transaction Agents [45.148328075418156]
  We study a deployed multi-turn food-and-beverage ordering agent and measure how many genuine quality problems it catches.<n>Our blind-spot taxonomy shows the failure is structured, not random.<n>For production multi-turn agents, automated judging is a regression floor, not a substitute for human review.
  arXiv  Detail & Related papers  (2026-06-09T02:11:01Z)
• VERDI: Single-Call Confidence Estimation for Verification-Based LLM Judges via Decomposed Inference [0.0]
  VERDI (VERification-Decomposed Inference Inference) is a method that extracts confidence from a structured judge.<n> VERDI decomposes each evaluation into sub-checks and derives three structural signals.<n>On three public benchmarks, VERDI achieves AUROC 0.72-0.91 on GPT-4.1-mini and 0.66-0.80 on GPT-5.4-mini.
  arXiv  Detail & Related papers  (2026-05-11T23:39:19Z)
• MEMSAD: Gradient-Coupled Anomaly Detection for Memory Poisoning in Retrieval-Augmented Agents [0.0]
  We formalize memory poisoning attacks on retrieval-augmented agents as a Stackelberg game with a unified evaluation framework.<n>We show faithful evaluation increases measured attack success by $4times$ (ASR-R: 0.25 to 1.00$)<n>Our primary contribution is MEMSAD, a calibration-based defense grounded in a gradient coupling.
  arXiv  Detail & Related papers  (2026-05-05T08:15:41Z)
• PHBench: A Benchmark for Predicting Startup Series A Funding from Product Hunt Launch Signals [0.0]
  Structured launch signals on Product Hunt contain statistically significant predictive information for Series A funding outcomes.<n>We construct PHBench from 67,292 featured Product Hunt posts spanning 2019-2025, linked to Crunchbase funding records via deterministic domain matching.<n>We identify 528 verified Series A raises within 18 months of launch (positive rate: 0.78%).
  arXiv  Detail & Related papers  (2026-05-03T17:03:33Z)
• ImplicitMemBench: Measuring Unconscious Behavioral Adaptation in Large Language Models [60.14219417402433]
  Existing memory benchmarks for LLM agents evaluate explicit recall of facts, yet overlook implicit memory where experience becomes automated behavior without conscious retrieval.<n>We introduce ImplicitMemBench, the first systematic benchmark evaluating implicit memory through three constructs drawn from standard cognitive-science accounts of non-declarative memory.<n>Our 300-item suite employs a unified Learning/Priming-Interfere-Test protocol with first-attempt scoring.
  arXiv  Detail & Related papers  (2026-04-09T10:26:32Z)
• Sparse Regression under Correlation and Weak Signals: A Reproducible Benchmark of Classical and Bayesian Methods [1.6679662639178268]
  We benchmark six sparse regression methods on synthetic data.<n>Bayesian methods win on prediction error (MSE 72 vs. 108-267), and the Horseshoe delivers near-nominal 95% coverage (94.8%)<n>For variable selection, Lasso and Spike-and-Slab tie at F1 0.47, making Lasso the practical default when posteriors are not needed.
  arXiv  Detail & Related papers  (2026-04-04T15:46:44Z)
• Learn by Surprise, Commit by Proof [0.0]
  We propose a self-gated post-training framework for autonomous knowledge acquisition.<n>When a passage produces anomalously high pertoken loss, L SCP flags it, generates a Q&A chain that forces the model to articulate its own knowledge.<n>The entire learning intensity is governed by a single parameter $r$.
  arXiv  Detail & Related papers  (2026-04-02T12:17:10Z)
• LLM Readiness Harness: Evaluation, Observability, and CI Gates for LLM/RAG Applications [51.56484100374058]
  We present a readiness harness for LLM and RAG applications that turns evaluation into a deployment decision workflow.<n>The system combines automated benchmarks, OpenTelemetry observability, and CI quality gates under a minimal API contract.<n>We evaluate the harness on ticket-routing and BEIR tasks with full Azure matrix coverage.
  arXiv  Detail & Related papers  (2026-03-28T18:03:32Z)
• TARG: Training-Free Adaptive Retrieval Gating for Efficient RAG [46.122203287541005]
  Training-free Adaptive Retrieval Gating (TARG) is a single-shot policy that decides when to retrieve using only a short, no-latency draft from the base model.<n>On NQ-Open, TriviaQA, and PopQA, TARG consistently shifts the accuracy-efficiency frontier.
  arXiv  Detail & Related papers  (2025-11-12T23:09:52Z)
• Lazy Layers to Make Fine-Tuned Diffusion Models More Traceable [70.77600345240867]
  A novel arbitrary-in-arbitrary-out (AIAO) strategy makes watermarks resilient to fine-tuning-based removal. Unlike the existing methods of designing a backdoor for the input/output space of diffusion models, in our method, we propose to embed the backdoor into the feature space of sampled subpaths. Our empirical studies on the MS-COCO, AFHQ, LSUN, CUB-200, and DreamBooth datasets confirm the robustness of AIAO.
  arXiv  Detail & Related papers  (2024-05-01T12:03:39Z)
• LLMs as Factual Reasoners: Insights from Existing Benchmarks and Beyond [135.8013388183257]
  We propose a new protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits. Most LLMs struggle on SummEdits, with performance close to random chance. The best-performing model, GPT-4, is still 8% below estimated human performance.
  arXiv  Detail & Related papers  (2023-05-23T21:50:06Z)
• Censored Quantile Regression Forest [81.9098291337097]
  We develop a new estimating equation that adapts to censoring and leads to quantile score whenever the data do not exhibit censoring. The proposed procedure named it censored quantile regression forest, allows us to estimate quantiles of time-to-event without any parametric modeling assumption.
  arXiv  Detail & Related papers  (2020-01-08T23:20:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.