DAILY PACE CARD — THE BLOG

News Radar — Monday, November 16, 2026

The daily News Radar digest for Monday, November 16, 2026: AI methods and news, and racing, handicapping and betting from the industry’s own outlets — scanned from free public sources, scored by a published rule, and linked to the original reporting.

The News Radar is the lab’s daily scan of free public sources, produced by AI. It reads two channels: AI methods and news — the tools behind the lab’s data work — and racing, handicapping and betting, drawn from the industry’s own outlets: Brisnet, Equibase, the Daily Racing Form, E-Ponies, Today’s Racing Digest and Thoroughbred Daily News, plus sportsbook, casino and lottery coverage. On Monday, November 16, 2026 it scanned 3672 items from 20 sources, kept 774 inside its 48-hour window, removed 123 duplicate(s), and kept 30 items. Every headline below belongs to its publisher and links to the original; the line under each item states the signal that kept it.

AI methods & news

AscendGYM: A Cost-Aware Kernel Optimization Benchmark for Real-World Workloads on Ascend NPUs

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.17621v1 Announce Type: new Abstract: Kernel optimization is essential for efficient large language model (LLM) training and inference on Ascend NPUs. Existing Ascend kernel benchmarks primarily focus on kernel generation, with less emphasis on the ability of agents to optimize existing…

Why it matters — matched AI methods signals: 'large language model', 'llm', 'benchmark', 'benchmarks'; source: arXiv cs.AI.

G-CAST: Governed Continual Self-Training for E-Commerce Agents under Evolving Demands

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.17814v1 Announce Type: new Abstract: With the rapid advancement of Large Language Models (LLMs), an increasing number of users are turning to LLM-powered agents for conversational shopping in e-commerce. However, the user needs continue to change as new product categories, procurement…

Why it matters — matched AI methods signals: 'llm', 'llms', 'agents', 'training'; source: arXiv cs.AI.

HARDTESTGEN: A High-Quality RL Verifier Generation Pipeline for LLM Algorithmic Coding

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2505.24098v2 Announce Type: cross Abstract: Verifiers provide important reward signals for reinforcement learning of large language models (LLMs). However, it is challenging to develop or create reliable verifiers, especially for code generation tasks. A well-disguised wrong solution program…

Why it matters — matched AI methods signals: 'llm', 'llms', 'reinforcement learning', 'code generation'; source: arXiv cs.AI.

Agent-written heuristics versus commercial solvers on unsolved job-shop benchmark instances

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.17682v1 Announce Type: cross Abstract: Commercial solvers are the standard tool for difficult scheduling problems. We study whether a frontier large language model, used as a coding agent with open-source tools only, can match them in quality and speed. We report every run, and…

Why it matters — matched AI methods signals: 'large language model', 'open-source', 'benchmark', 'agent'; source: arXiv cs.AI.

Superfluid: Local AI Server for Distributed LLM Inference

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.17753v1 Announce Type: cross Abstract: Running large language models (LLMs) on local devices keeps data private, removes per-token cost, cuts network latency, and keeps working without a connection. Two shifts now strain how such models are served. On one device, several agents share a…

Why it matters — matched AI methods signals: 'llm', 'llms', 'inference', 'agents'; source: arXiv cs.AI.

Can Large Language Models Optimize Pretrained Neural Network Weights? A Multi-Phase Empirical Evaluation of Gradient-Free Editing in YOLO26 Object Detection

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.17789v1 Announce Type: cross Abstract: Direct weight editing offers a gradient-free alternative to fine-tuning pretrained detectors, but its generalization and the value of large language model (LLM) guidance remain uncertain. We evaluate five experimental phases using YOLO26 models and…

Why it matters — matched AI methods signals: 'large language model', 'llm', 'fine-tuning', 'neural'; source: arXiv cs.AI.

MineralExp: Benchmarking LLM Agents for Experimental Design in Real-World Mineral Exploration

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.17931v1 Announce Type: cross Abstract: In mineral exploration, verifying geological hypotheses relies on designing a variety of geochemical experiments. While large language model (LLM) agents increasingly automate downstream geoscientific data analysis, a crucial upstream capability…

Why it matters — matched AI methods signals: 'large language model', 'llm', 'agents'; source: arXiv cs.AI.

From Chats to Markets: AgenticPay for LLM-Powered Negotiation in Multi-Agent Commerce

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2602.06008v2 Announce Type: replace Abstract: Large language model agents are increasingly expected to negotiate and transact autonomously, yet existing benchmarks rarely combine multimodal product grounding, multi-issue bargaining, and competitive multi-agent markets. We introduce…

Why it matters — matched AI methods signals: 'large language model', 'llm', 'benchmarks', 'agent'; source: arXiv cs.AI.

DeltaBox: Millisecond-Scale Sandbox Checkpoint and Rollback for Stateful AI Agents

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2605.22781v3 Announce Type: replace-cross Abstract: Emerging LLM-powered AI agent workloads, such as test-time tree search and reinforcement learning (RL), require high-frequency state exploration, relying on rapid checkpoint/restore (C/R) of the complete sandbox state, including files and…

Why it matters — matched AI methods signals: 'llm', 'ai agent', 'agent', 'agents'; source: arXiv cs.AI.

Agentic-OPD: Recovering Forgotten Agent Capabilities via On-Policy Distillation for Agentic Tasks

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.12799v2 Announce Type: replace-cross Abstract: Reinforcement learning is now the mainstream paradigm for training LLMs on agentic tasks, yet standard practice keeps only the best-validation checkpoint and discards the rest. We show that this is lossy: in GRPO training of LLM agents, the…

Why it matters — matched AI methods signals: 'llm', 'llms', 'agent', 'agents'; source: arXiv cs.AI.

JudgeMix: Judge-Guided Mixed-Rollout Data Synthesis for LLM Agents

arXiv cs.AI · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.16174v2 Announce Type: replace-cross Abstract: Supervised fine-tuning (SFT) on expert trajectories is a standard paradigm for training LLM agents. However, it suffers from a well-known training-inference distribution mismatch, calling for target-aware data synthesis strategies that…

Why it matters — matched AI methods signals: 'llm', 'inference', 'agents', 'fine-tuning'; source: arXiv cs.AI.

TIDE: Reclaiming Lost Performance In On-Policy Reinforcement Learning

arXiv cs.LG · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.17924v1 Announce Type: new Abstract: Synchronous, colocated RL post-training of large language models (LLMs) today suffers from severe under-utilization due to long-tail rollouts. Inference engines fall idle waiting for long-tail rollouts or decode at low, inefficient batch sizes. To…

Why it matters — matched AI methods signals: 'llms', 'inference', 'training', 'reinforcement learning'; source: arXiv cs.LG.

ArchSpace: Recording, Reviewing, and Referencing LLM Architecture Research

arXiv cs.LG · Nov 16, 12:00 AM ET · score 10.5

arXiv:2611.17332v2 Announce Type: replace-cross Abstract: Model architecture shapes the scaling behavior, training efficiency, and inference cost of LLMs, making it a central focus of both industry and academia. In industry, teams such as those behind DeepSeek, Qwen, and Kimi routinely introduce…

Why it matters — matched AI methods signals: 'llm', 'llms', 'inference', 'training'; source: arXiv cs.LG.

More Agents, More Samples, or More Thinking? Untangling Coordination, Ensembling, and Test-Time Compute in Multi-Agent LLMs

arXiv cs.AI · Nov 16, 12:00 AM ET · score 9.5

arXiv:2611.17610v1 Announce Type: new Abstract: Evaluations of multi-agent language models often compare communicating architectures against a single greedy pass. This practice confounds coordination with sample multiplicity and inference compute. We synthesize nine primary empirical evaluations…

Why it matters — matched AI methods signals: 'llms', 'inference', 'agent', 'agents'; source: arXiv cs.AI.

Aligning AI Teams

arXiv cs.AI · Nov 16, 12:00 AM ET · score 9.5

arXiv:2611.17820v1 Announce Type: new Abstract: Previous work has shown that alignment training on individual LLM agents can fail to transfer to multi-agent settings. We investigate the underlying mechanisms for this failure, and explore interventions to preserve alignment transfer to multi-agent…

Why it matters — matched AI methods signals: 'llm', 'agent', 'agents', 'training'; source: arXiv cs.AI.

Racing, handicapping & betting

'You're looking at a full-on casino and sports betting': Orr says… - 1819 News

Google News: casino gambling · Nov 15, 03:05 PM ET · score 8.0

'You're looking at a full-on casino and sports betting': Orr says… 1819 News

Why it matters — matched gambling signals: 'sports betting', 'casino'; source: Google News: casino gambling.

After Another Move to Crush the Preakness, Will Maryland Fight Back?

Thoroughbred Daily News · Nov 15, 02:26 PM ET · score 7.0

The Triple Crown was already finished. That became official when NYRA and Churchill Downs created the Thoroughbred Championship Series (TCS), a rich, bonus-laden six-race series that includes the GI Kentucky Derby and the GI Belmont Stakes, but not the GI Preakness. That left the Preakness alone on…

Why it matters — matched gambling signals: 'thoroughbred', 'stakes'; source: Thoroughbred Daily News.

‘You Can Really Make A Killing’ – Buyer And Seller Insight From Two Major Players Ahead Of The Goffs November Foal Sale

Thoroughbred Daily News · Nov 15, 01:09 PM ET · score 7.0

Vinesgrove Stud boss Billy McEnery and leading talent scout Alex Butler provide a buyer versus seller insight ahead of the Goffs November Foal Sale Horse Racing Ireland ran an advertorial campaign in the build-up to the Irish Champions Festival with the tagline, “it all comes down to this.” That…

Why it matters — matched gambling signals: 'horse racing'; source: Thoroughbred Daily News.

Handicapping: How Auction Prices Relate to Race Results - BloodHorse

Google News: horse racing handicapping · Nov 16, 06:02 AM ET · score 7.0

Handicapping: How Auction Prices Relate to Race Results BloodHorse

Why it matters — matched gambling signals: 'handicapping'; source: Google News: horse racing handicapping.

DRF's Horse Racing Playbook for Sunday, Nov. 15, 2026 - Daily Racing Form

Google News: Daily Racing Form · Nov 15, 07:39 AM ET · score 6.5

DRF's Horse Racing Playbook for Sunday, Nov. 15, 2026 Daily Racing Form

Why it matters — matched gambling signals: 'horse racing'; source: Google News: Daily Racing Form.

State senator files bill to give $1,500 to Texas families, argues casino gambling and cannabis taxes could fund it - CDC Gaming

Google News: casino gambling · Nov 15, 07:48 PM ET · score 6.5

State senator files bill to give $1,500 to Texas families, argues casino gambling and cannabis taxes could fund it CDC Gaming

Why it matters — matched gambling signals: 'casino', 'gambling'; source: Google News: casino gambling.

Man o’ War Among De Paz Options for Sounds Like a Plan

Thoroughbred Daily News · Nov 15, 03:01 PM ET · score 6.0

Club Sixty Five Racing's Sounds Like a Plan ( Twirling Candy ) could mark his return in the GII Man o' War Stakes, a 10-furlong turf route for 3-year-olds and up set for Saturday, Nov. 28, at Belmont Park. Trainer Horacio De Paz has nominated Sounds Like a Plan to both the Man o' War and the The…

Why it matters — matched gambling signals: 'trainer', 'stakes'; source: Thoroughbred Daily News.

Brown’s Nevada, Election Hero and Bletchley Possible for Remsen

Thoroughbred Daily News · Nov 15, 01:54 PM ET · score 6.0

A trio of recent eye-catching maiden winners at Belmont Park for trainer Chad Brown–Nevada (Uncle Mo), Election Hero ( Early Voting ) and Bletchley ( Gun Runner )–are under consideration for the GII Remsen Stakes on Dec. 5. The nine-furlong test for juveniles is part of the Road to the Kentucky…

Why it matters — matched gambling signals: 'trainer', 'stakes'; source: Thoroughbred Daily News.

Stewart Manor Winner Pink Venom Joins Amelia Green Barn, to Prepare for U.S. Campaign

Thoroughbred Daily News · Nov 15, 01:39 PM ET · score 6.0

Team Valor International's Pink Venom (Ire) (Soldier's Call {Ire}) has settled into the Belmont Park barn of trainer Amelia Green following her last-to-first triumph for previous trainer Johnny Murtagh in Saturday's $175,000 Stewart Manor Stakes, a six-furlong sprint for juvenile fillies slated for…

Why it matters — matched gambling signals: 'trainer', 'stakes'; source: Thoroughbred Daily News.

Mountaineer Casino Racetrack & Resort | Nov 15, 2026 - Equibase

Google News: Equibase · Nov 15, 09:53 PM ET · score 6.0

Mountaineer Casino Racetrack & Resort | Nov 15, 2026 Equibase

Why it matters — matched gambling signals: 'casino'; source: Google News: Equibase.

Mountaineer Casino Racetrack & Resort Index - Equibase

Google News: Equibase · Nov 15, 08:02 PM ET · score 6.0

Mountaineer Casino Racetrack & Resort Index Equibase

Why it matters — matched gambling signals: 'casino'; source: Google News: Equibase.

Gambling Regulation Roundup: Louisiana, CFTC, Vietnam - Casino News Daily

Google News: casino gambling · Nov 15, 05:23 PM ET · score 6.0

Gambling Regulation Roundup: Louisiana, CFTC, Vietnam Casino News Daily

Why it matters — matched gambling signals: 'casino', 'gambling'; source: Google News: casino gambling.

Sports Betting States Where It’s Legal in the US 2026 (Full List) - FOX Sports

Google News: sports betting · Nov 15, 02:55 PM ET · score 6.0

Sports Betting States Where It’s Legal in the US 2026 (Full List) FOX Sports

Why it matters — matched gambling signals: 'sports betting'; source: Google News: sports betting.

Horse Racing Tips: Betfred's NAP of the Day for Monday 16 November 2026 - Betfred Insights

Google News: horse racing handicapping · Nov 15, 01:08 PM ET · score 6.0

Horse Racing Tips: Betfred's NAP of the Day for Monday 16 November 2026 Betfred Insights

Why it matters — matched gambling signals: 'horse racing'; source: Google News: horse racing handicapping.

UK gambling trade bodies call for more apprenticeships - Casino.com

Google News: casino gambling · Nov 15, 12:13 PM ET · score 6.0

UK gambling trade bodies call for more apprenticeships Casino.com ---

Why it matters — matched gambling signals: 'casino', 'gambling'; source: Google News: casino gambling.

How the radar works

The radar scores every item by a published rule: a source weight, weighted keyword signals matched as exact token sequences, a recency score, and a Hacker News points bonus. It keeps the top 15 items in each channel. The full rule and the source list are documented in the radar’s project notes. Google News links are Google’s redirect URLs, exactly as its feeds provide them; they open the publisher’s article in a browser.

Sources scanned this run: Hacker News, arXiv cs.AI, arXiv cs.LG, Hugging Face Blog, OpenAI News, Google Research Blog, Simon Willison, GitHub Trending (RSS mirror), Thoroughbred Daily News, arXiv stat.AP, Google News: horse racing handicapping, Google News: Brisnet, Google News: Equibase, Google News: Daily Racing Form, Google News: E-Ponies, Google News: Today's Racing Digest, Google News: horse racing workouts, Google News: sports betting, Google News: casino gambling, Google News: lottery statistics.

Outlets that offer no working free feed from this network — the Paulick Report and BloodHorse among them — are not scanned; nothing is carried from a source the radar cannot read.