Trevor French
@TrvrFrnch
Followers
1K
Following
5K
Media
89
Statuses
345
Data Scientist | PhD Student | On a mission to become the AI Hokage
North Carolina, USA
Joined August 2021
Use code “whitehouse” at checkout.
🇺🇸@Ford will offer employee pricing to all US shoppers:. "'For 121 years, we've put our money where our mouth is, assembling vehicles that Americans rely on and supporting American jobs,' Ford said in a statement."
0
0
1
Cognitive Security (COGSEC) studies how online influence shapes minds at scale (Waltzman, 2017). Yet, individuals must also act—practicing original thinking can safeguard our cognitive space.
1
0
2
One of my favorite albums is @DGDtheband’s “Jackpot Juicer”. As a data scientist, I couldn’t resist digging deeper—so I ran each song through Librosa, a Python library for audio analysis, to see what the numbers could tell me about this record.
2
0
2
Flags acquired ✅
0
0
2
I’m looking for inspiration for my portfolio website. Send me your favorite ones (or your own)!.
0
0
2
RT @_mario_neo_: I built my own PyTorch from scratch over the last 5 months in C and modern Python. Check it out on GitHub: .
github.com
(WIP) A small but powerful, homemade PyTorch from scratch. - MarioSieg/magnetron
0
167
0
(7/7) "Show Your Work: Scratchpads for Intermediate Computation with Language Models" by Nye et al. (2021) introduces a framework for models to simulate code execution by reading from and writing to a “scratchpad,” where intermediate calculations are recorded.
0
0
1
(6/7) "Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks" by W. Chen et al. (2022) presents PoT, a strategy that requires the LLM to perform reasoning but passes computation off to an interpreter.
arxiv.org
Recently, there has been significant progress in teaching language models to perform step-by-step reasoning to solve complex numerical reasoning tasks. Chain-of-thoughts prompting (CoT) is by far...
1
0
1
(5/7) "Chain of Code: Reasoning with a Language Model-Augmented Code Emulator" by Li et al. (2023) puts forth CoC, which allows models to reason through code that is executed through either an interpreter or an "LMulator." .
1
0
0
(4/7) "Self-Refine: Iterative Refinement with Self-Feedback" by Madaan et al. (2023) presents Self-Refine, a prompting technique where a model iteratively provides itself with feedback to refine subsequent responses.
arxiv.org
Like humans, large language models (LLMs) do not always generate the best output on their first try. Motivated by how humans refine their written text, we introduce Self-Refine, an approach for...
1
0
0