@rquant.bsky.social
If only we had software to handle such requests… something that could search the internet for us, check the references, etc. Sigh…
August 25, 2025 at 1:20 PM
If only we had software to handle such requests… something that could search the internet for us, check the references, etc. Sigh…
Why are chain of thought models so good at math?
May 19, 2025 at 8:19 AM
Why are chain of thought models so good at math?
Apologies for asking it again here, but really curious about your view on this.
April 29, 2025 at 4:09 PM
Apologies for asking it again here, but really curious about your view on this.
A good faith question:If you believe LLMs are essentially useless,how do you explain their significant progress on math olympiad problems?These questions are extremely difficult and require serious reasoning skills(and if we don’t call IMO-level problems reasoning, not sure what else would qualify)
April 29, 2025 at 4:07 PM
A good faith question:If you believe LLMs are essentially useless,how do you explain their significant progress on math olympiad problems?These questions are extremely difficult and require serious reasoning skills(and if we don’t call IMO-level problems reasoning, not sure what else would qualify)
A good faith question:If you believe LLMs are essentially useless,how do you explain their significant progress on math olympiad problems?These questions are extremely difficult and require serious reasoning skills(and if we don’t call IMO-level problems reasoning, not sure what else would qualify)
April 26, 2025 at 9:15 PM
A good faith question:If you believe LLMs are essentially useless,how do you explain their significant progress on math olympiad problems?These questions are extremely difficult and require serious reasoning skills(and if we don’t call IMO-level problems reasoning, not sure what else would qualify)