/Tech6h ago

Study Shows Large Reasoning Models Ignore Instructions During Reasoning

120852.8K

#753

Original post

James Zou@james_y_zou#753inTech

We show that LLMs are bad at following instructions while reasoning. Will be presented at #ACL2026 this week!

Together AI@togethercompute

🧠Do reasoning models really follow our instructions?

Together AI’s newest paper "ReasonIF: Large Reasoning Models Fail to Follow Instructions During Reasoning", studies how well large reasoning models (LRMs) follow user instructions during reasoning.

Authors: @ykwon_0407, @ShangZhu18, @federicobianchy, @KaitlynZhou, @james_y_zou

9:02 AM · Jul 2, 2026 · 2.3K Views

Sentiment

Sentiment building, check back later.

Cluster Engagement

Digg Deeper

No Digg Deeper questions have been answered for this story yet.

Posts from X

Most Activity

VIEWS67RETWEETS2

Yongchan Kwon@ykwon_0407

Are frontier LLMs faithfully following user instructions?

We investigated how well they adhere to instructions within their reasoning traces, and @ShangZhu18 will be presenting ReasonIF at #ACL2026!

Where? Poster Session C (Grand Hall) When? Sunday, July 5, 4:00–5:30 PM

James Zou@james_y_zou

We show that LLMs are bad at following instructions while reasoning. Will be presented at #ACL2026 this week!

4h41951