Scoop watch

A real example: what you get

Example in Astrophysics Computer science and AI Biology and medicine

15 new papers close to the project “Grounding Answers in Retrieved Passages: A Study of Hallucinations in Retrieval-Augmented Language Models”, found in the announcements of its first days

How close it is on each facet: question, method, data, results
What it shares with your project and what stays your own
What to do: cite it, set your project apart, contact the authors, hurry
The prior art of the last three years, and the new papers of the teams you name

Partial overlap

PCG-RAG addresses the risk that partially supported entity hypotheses redirect retrieval toward the wrong evidence. This is closely related to the project's finding that retrieval loses its grounding benefit when passages concern a related but different entity.

PCG-RAG: Progressive Compositional Grounding for Retrieval-Augmented Generation in Complex Web Question Answering

What stays your own

The project measures unsupported answer sentences using NLI and compares retrieval with a generator without retrieval; PCG-RAG instead proposes a multi-stage framework and reports QA accuracy. Emphasize the controlled measurement of entity-mismatched retrieval and the separation of retrieval failures from generator failures.

The email you would have received · 15 new papers are close to your projects

The first results

PCG-RAG: Progressive Compositional Grounding for Retrieval-Augmented Generation in Complex Web Question Answering Partial overlap PCG-RAG addresses the risk that partially supported entity hypotheses redirect retrieval toward the wrong evidence. This is closely related to the project's finding that retrieval loses its grounding benefit when passages concern a related but different entity.
Search-by-Answer Bias in Retrieval Models Partial overlap This paper identifies a dense-retrieval failure in which familiar answer objects receive favorable rankings even in passages expressing an incorrect relation. It provides a related account of how retrieval can select misleading evidence rather than protect against unsupported answers.

13 more results hidden

See a complete example, explained step by step →

Every example, in your field →

Create a free account, then Premium runs it on your own work. Continue with Google Create a free account Log in

How it works

Describe your project in a few paragraphs (question, method, data, expected results), or upload the PDF of your draft: an AI writes the description from it.
At once, the papers of the last 3 years are searched on arXiv, in the journals and on OpenReview, with those that already cite your key references: you see what is already published close to your project.
Every day, the new papers are compared with each facet of your project apart (question, method, data, results), and with the names that identify it. An AI reads the closest abstracts.
When one overlaps your project, you get an email: a strong overlap (same question, same or very close approach) or a partial one, with how close it is on each facet.
For each paper, the AI says what it shares with your project, what stays yours and what to do: cite it, set your project apart, contact the authors or publish soon. A strong overlap is compared with the full text of the paper when its license allows it.

Follow competing teams, and give your key references: their new papers, and the papers that cite your references, are always read.
Mark a paper as not relevant and similar ones are pushed down; add the close ones to a collection.
Your own papers are never reported: give your name and those of your usual coauthors.
Up to 5 projects watched.
Your description stays private and is only used for the comparison. A PDF is read once and is not stored, unless you ask us to keep its text for the full-text comparisons.

Useful during a thesis or a long project, to know early that a competing team is close, cite it, and adjust how you present your work.