agential
agential copied to clipboard
agential-ai
🔔🧠 Easily experiment with popular language agents across diverse reasoning/decision-making benchmarks!
### Feature Description HumanEval has a robust code interpreter set up that uses the Python `exec()` function internally. We currently just have `safe_execute` which doesn't have all the guardrails in-place...
**Implement**: - [x] HotpotQA - [x] TriviaQA - [x] #87 - [x] #71 - [x] #73 - [x] #72 - [x] #80 - [x] #81 - [x] #83 - [...
### Feature Description The Reflexion implementation only has the supported [prompts](https://github.com/agential-ai/agential/blob/main/agential/cog/prompts/reflexion.py). Add relevant prompts and logic to the current Reflexion implementation. You'll see that an agent's current structure is divided...
### Feature Description The Reflexion implementation only has the supported [prompts](https://github.com/agential-ai/agential/blob/main/agential/cog/prompts/reflexion.py). Add relevant prompts and logic to the current Reflexion implementation. You'll see that an agent's current structure is divided...
### Feature Description The Reflexion implementation only has the supported [prompts](https://github.com/agential-ai/agential/blob/main/agential/cog/prompts/reflexion.py). Add relevant prompts and logic to the current Reflexion implementation. You'll see that an agent's current structure is divided...
### Feature Description Familiarize yourself with the repository and take a look at the ReAct [repo](https://github.com/ysymyth/ReAct), [paper](https://arxiv.org/abs/2210.03629), and [AgentBench](https://arxiv.org/abs/2308.03688). Currently, the ReAct implementation only has [prompts](https://github.com/alckasoc/agential/blob/main/agential/cog/prompts/react.py) for HotpotQA and FEVER....
### Feature Description Familiarize yourself with the repository and take a look at the ReAct [repo](https://github.com/ysymyth/ReAct), [paper](https://arxiv.org/abs/2210.03629), and [WebShop](https://arxiv.org/abs/2207.01206). Currently, the ReAct implementation only has [prompts](https://github.com/alckasoc/agential/blob/main/agential/cog/prompts/react.py) for HotpotQA and FEVER....
### Feature Description Familiarize yourself with the repository and take a look at the ReAct [repo](https://github.com/ysymyth/ReAct), [paper](https://arxiv.org/abs/2210.03629), and [ALFWorld](https://alfworld.github.io/). Currently, the ReAct implementation only has [prompts](https://github.com/alckasoc/agential/blob/main/agential/cog/prompts/react.py) for HotpotQA and FEVER....
### Feature Description Familiarize yourself with the repository and take a look at the CRITIC [repo](https://github.com/microsoft/ProphetNet/tree/master/CRITIC), [paper](https://arxiv.org/abs/2305.11738), and [WebShop](https://arxiv.org/abs/2207.01206). Currently, the CRITIC implementation only has [prompts](https://github.com/alckasoc/agential/blob/main/agential/cog/prompts/critic.py) for HotpotQA and TriviaQA....
### Feature Description Familiarize yourself with the repository and take a look at the CRITIC [repo](https://github.com/microsoft/ProphetNet/tree/master/CRITIC), [paper](https://arxiv.org/abs/2305.11738), and [WebShop](https://arxiv.org/abs/2207.01206). Currently, the CRITIC implementation only has [prompts](https://github.com/alckasoc/agential/blob/main/agential/cog/prompts/critic.py) for HotpotQA and TriviaQA....
Metadata
Owner
Metadata
🔔🧠 Easily experiment with popular language agents across diverse reasoning/decision-making benchmarks!