Abel_TM
@abeltm.bsky.social
Research Scientist. Implementing reasoning in AI. Theory and implementation of open ended reasoning algorithms for long term planning, robotics, math, protein design and science
I would put a like but was stopped by the perfect number...
December 10, 2024 at 9:03 AM
I would put a like but was stopped by the perfect number...
Eg: "Find a possible sequence of movements from the start of a game of chess that leads to white pieces delivering checkmate in four moves. Only knights and pawns can be moved"
- GPT(4o, o1-mini, o1-preview): Impossible
- Gemini-1.5-Pro-002: 1. Nf3 Nf6 2. Ng1 Ng8 3. f4 e5 4. g4 h5# ???
- Claude:
- GPT(4o, o1-mini, o1-preview): Impossible
- Gemini-1.5-Pro-002: 1. Nf3 Nf6 2. Ng1 Ng8 3. f4 e5 4. g4 h5# ???
- Claude:
December 5, 2024 at 10:31 AM
Eg: "Find a possible sequence of movements from the start of a game of chess that leads to white pieces delivering checkmate in four moves. Only knights and pawns can be moved"
- GPT(4o, o1-mini, o1-preview): Impossible
- Gemini-1.5-Pro-002: 1. Nf3 Nf6 2. Ng1 Ng8 3. f4 e5 4. g4 h5# ???
- Claude:
- GPT(4o, o1-mini, o1-preview): Impossible
- Gemini-1.5-Pro-002: 1. Nf3 Nf6 2. Ng1 Ng8 3. f4 e5 4. g4 h5# ???
- Claude:
Interesting results on reasoning potential with LLMs. I use regularly chess to test reasoning abilities and they usually ‘hallucinate’ invalid moves and positions.
From my work on general reasoning agents I see two main required properties: accuracy and flexibility.
From my work on general reasoning agents I see two main required properties: accuracy and flexibility.
December 5, 2024 at 10:31 AM
Interesting results on reasoning potential with LLMs. I use regularly chess to test reasoning abilities and they usually ‘hallucinate’ invalid moves and positions.
From my work on general reasoning agents I see two main required properties: accuracy and flexibility.
From my work on general reasoning agents I see two main required properties: accuracy and flexibility.