[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

![neural_avb Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::1754194661084983296.png) AVB [@neural_avb](/creator/twitter/neural_avb) on x 2133 followers
Created: 2025-07-20 11:26:00 UTC

X things I am looking forward in LLM Reasoning research:

X. Applying RL on open-ended tasks, instead of just strictly verifiable tasks

X. A self-supervised way to assign credits to individual/group of tokens given a response

X. Advantage vanishing issues as model gets smart


XXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1946894354653610327/c:line.svg)

**Related Topics**
[llm](/topic/llm)

[Post Link](https://x.com/neural_avb/status/1946894354653610327)