Introducing Interwhen: Steering reasoning agents with real-time verification
170
Next
35 days – 3067:40Popular
35 days – 61413:34
Published on 14 May 2026, 17:06
What if AI agents could check their work as they go? This verification method extracts verifiable properties from natural language and evaluates them using symbolic or model-based verifiers. Interwhen, a new open-source library, enables real-time verification of each step, helping agents act more safely and reliably in complex, real-world tasks.
Paper: arxiv.org/abs/2602.11202
GitHub: github.com/microsoft/interwhen
This session aired on May 14, 2026, at Microsoft Research Forum, Season 2 Episode 4.
Register for the series to hear about new releases: microsoft.com/en-us/research/e...
Explore all previous episodes: aka.ms/researchforumYTplaylist
Paper: arxiv.org/abs/2602.11202
GitHub: github.com/microsoft/interwhen
This session aired on May 14, 2026, at Microsoft Research Forum, Season 2 Episode 4.
Register for the series to hear about new releases: microsoft.com/en-us/research/e...
Explore all previous episodes: aka.ms/researchforumYTplaylist
Fresh videos
2 days – 7551:57
2 days – 1 1310:13
5 days – 558 3480:16
5 days – 15 2590:34
6 days – 35 49014:09
6 days – 59 7530:10Random videos
158 days – 12 2363:08
304 days – 1 5150:34
03.03.25 – 1 8384:23
04.07.24 – 1 052 37915:51
14.12.23 – 4720:47
20.06.08 – 25 5751:28
1 day – 1 4751:38
1 day – 1 3631:51
1 day – 4 9585:32
6 days – 2 2100:30
7 days – 3 0760:16
8 days – 7 9200:05
8 days – 16220:27
