Your model already knows the answer: how benchmark answers leak into LLMs
The contamination problem in AI model testing refers to the issue of models being able to recall answers from their training data rather than reasoning their way to a solution. This problem arises when benchmarks are built from resolved events with public outcomes, allowing models to recognize and recall the answers. As a result, high scores on these benchmarks do not necessarily indicate a model's ability to reason, but rather its ability to recall previously learned information.
- ▪The contamination problem is a significant issue in AI model testing, where models can recall answers from their training data rather than reasoning their way to a solution.
- ▪Benchmarks built from resolved events with public outcomes are particularly susceptible to this problem, as models can recognize and recall the answers from their training data.
- ▪There are three main routes through which answers can leak into models: input leak, benchmark leak, and outcome leak, each with established basis in research.
Hacker News (Front Page) files mainly under programming. We currently carry 1,193 of its stories. Top-voted stories on Hacker News.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Elman |
| Canonical URL | https://elman.ai/news/your-model-already-knows-the-answer/ |
| Publication time | Wed, 05 Aug 2026 16:56:56 +0000 |
| Retrieval time | 2026-08-05T19:05:47.442Z |
| Last seen | 2026-08-05T19:05:47.442Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | ANO2NEqNSx-_ · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Essay · 30 July 2026 Your Model Already Knows the Answer Francesco Moramarco • Co-founder, Elman The best test of an AI model is a real world problem with a known answer. It is also the easiest test to cheat. An AI model has read much of the internet, so it may have met your question, the answer or both long before you asked. A high score then hides two very different things: a model that reasoned its way to the answer, and one that is repeating something it already read. Telling those apart is the contamination problem, and it sits under a surprising share of the benchmarks the field relies on. Imagine you want to know how reliably a model can predict whether a drug will succeed in the clinic.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Elman.
Discussion
0 commentsMore from Elman
