Cover image

When a Failed Workflow Still Produces a Learning Signal

TL;DR for operators An automation agent may retrieve the right information, update one application, and complete several intermediate operations before failing its final task. A binary success test assigns that entire attempt zero reward, even when parts of the workflow were executed correctly. Explorer-Zero1 demonstrates a training alternative: let one agent generate achievable tasks through actual environment interaction, record verifiable intermediate states, and let another agent learn from those checkpoints. ...

November 2, 2026 · 8 min · Zelina
Cover image

The Right Tool Is Not the Whole Workflow

TL;DR for operators An AI agent can select the correct function and still fail the overall workflow because an earlier result was lost, a prerequisite was skipped, or the wrong patient or treatment-plan identifier was passed downstream. RadOnc-Agent makes these risks measurable. In matched synthetic tests, preserving longitudinal workflow state increased completion from 84.00% to 96.50%. A separate validation experiment found that deterministic checks prevented all deliberately mismatched backend dispatches, whereas bypassing those checks allowed 95.28% to proceed through restricted replay/test interfaces. ...

November 1, 2026 · 8 min · Zelina
Cover image

The Agent Scored Higher. What Actually Improved?

TL;DR for operators An AI agent can perform remarkably well on held-out benchmark questions without reliably solving the intended task. On LiveMath, an evolved agent answered all 53 held-out questions containing a recurring answer option correctly, but only 22 of the other 71. The recurring pattern provided a shortcut that survived into the supposedly independent evaluation. ...

October 31, 2026 · 8 min · Zelina
Cover image

When the Newest AI Skill Isn't the Best

TL;DR for operators An AI agent can fix one visual-reasoning failure and introduce another. As its instructions and tools evolve, operators need a way to distinguish genuine capability improvements from revisions that merely solve recent examples. They also need to diagnose recurring failures without repeatedly examining entire execution histories. VisualEvolve addresses both problems through versioned skill libraries and structured evidence retrieval. Across four frozen vision-language models and six evaluation sets, it improves over the initial skill library in 23 of 24 comparisons, with one tie. Removing historical version selection or structured evidence access substantially reduces performance. For organizations maintaining AI agents, the research supports treating skill revisions as evaluated, reversible artifacts rather than automatically accepting the latest changes. It does not yet establish enterprise reliability or lower total operating costs. ...

October 31, 2026 · 8 min · Zelina
Cover image

The Retry Is Not the Lesson: How ReTeach Learns from Failed Answers

TL;DR for operators A model generates an incorrect answer. An automated verifier can reject it but cannot explain the mistake. ReTeach1 shows how that failure can become training material without a worked solution or a larger external teacher. The model’s teacher revisits unsuccessful attempts, reflects on possible errors, and retries. But the successful retry is not simply copied into the student’s training targets. Instead, the teacher uses its additional reasoning context to produce better-informed token predictions along the student’s original response. The student learns from those predictions without needing that context itself. On SciKnowEval-Physics, explicit reflection achieves 64.24% accuracy, versus 62.91% with retries but no reflection (Section 4.3, Figure 4). ...

October 30, 2026 · 8 min · Zelina
Cover image

A Shorter Prompt Is Only as Safe as Its Test

TL;DR for operators Production-agent prompts can be shortened without a measurable decline in benchmark correctness, but the benchmark must first be trustworthy. A documented optimization failure removed important instructions while passing sampled checks, exposing a risk that aggregate scores can conceal. A new study addresses this through reproducible evaluation of agents connected to changing business data, followed by correctness-gated prompt optimization. Across three model families, it reports 3.5–7.6% reductions in the complete instruction surface, with a statistically significant correctness improvement for one family. ...

October 29, 2026 · 8 min · Zelina
Cover image

Verified Is Not Authorized: How Agent Guardrails Earn the Right to Act

TL;DR for operators An agent guardrail that learns from incidents must solve three different problems: produce a better rule, verify that the rule is executable and internally consistent, and establish enough evidence to let it control real actions. VEPA separates these responsibilities. An LLM proposes policy changes, a deterministic verifier rejects malformed or conflicting rules, and a Bayesian authority mechanism determines which verified rules may actively allow or refuse agent actions. ...

October 28, 2026 · 8 min · Zelina
Cover image

The Search Stack Changes the Answer

TL;DR for operators An AI research assistant’s accuracy cannot be reliably predicted by selecting the strongest language model, connecting a powerful search service, and adding specialized agents. Those components must work together. In HyperBrowseComp1, replacing native search with Exa improved GPT-5.6 Sol accuracy from 19.15% to 26.71%, while reducing Gemini 3.7 Flash accuracy from 31.68% to 22.22%. A separate multi-agent configuration completed only 330 of 423 attempts and answered 17.73% correctly overall. ...

October 27, 2026 · 8 min · Zelina
Cover image

Found but Unread: Why AI Agents Recommend the Websites They Can Access

TL;DR for operators A business can appear in AI search results without supplying the information an agent needs to recommend it. Across 1,056 matched businesses, websites classified as agent-ready received clear recommendations in 20.4% of agent journeys, versus 10.6% for less accessible websites. First-party-grounded answers were also more frequent: 77.8% versus 55.5%. ...

October 24, 2026 · 9 min · Zelina
Cover image

Better Predictions, Fewer Ideas: The Alpha Agent's Exploration Problem

TL;DR for operators An automated quantitative researcher can improve its predictions while becoming less effective at discovering different sources of predictive information. Better factor scores, more formulas, and broader productive research are not interchangeable measures of success. AlphaDiverse addresses this problem by training local agents to select complementary economic hypotheses and implement them as executable factors. Its reinforcement-learning objective rewards both predictive improvement and contributions that existing signals do not already provide. ...

October 23, 2026 · 10 min · Zelina