batterylake

6 ms·

1.
by batterylake 6mo ago
Very insightful! I found the section on killing papers to be a helpful reminder. As a Ph.D. student, this can be particularly challenging as your environment expects somewhat steady progress (annual reviews, advisor meetings, etc.), and you
2.
Codex Is a Surprisingly Competent Research Assistant(nicholasclark.org)
3 points by batterylake 7mo ago|0 comments
3.
by batterylake 11mo ago
Very interesting topic. I also wonder why other signs of AI writing, such as negative parallelism ("It's not just X, it's Y"), are preferred by the models. Also, I wrote a small extension that automatically replaces Chat
4.
by batterylake 11mo ago
Thanks!
5.
Show HN: I stopped ChatGPT from using em dashes(github.com)
6 points by batterylake 11mo ago|3 comments
6.
by batterylake 11mo ago
Who said Europe can't innovate
7.
Epistemic Alignment: A Mediating Framework for User-LLM Knowledge Delivery(arxiv.org)
2 points by batterylake 1y ago|0 comments
8.
by batterylake 2y ago
Cool, thanks!
9.
by batterylake 2y ago
Hi Claude Code team, excited for the launch! How well does Claude Code do on tasks which rely heavily on visual input such as frontend web dev or creating data visualizations?
10.
by batterylake 2y ago
I found it quite insightful and well done. Curious if this type of documentation will become more common like you mentioned.
11.
OpenAI Model Spec(model-spec.openai.com)
12 points by batterylake 2y ago|4 comments
12.
by batterylake 2y ago
I agree, it kind of reminds me of this paper that shows LLMs, just like humans, will preferentially remember information, which can lead to biased outputs. https://www.pnas.org/doi/abs/10.1073/pnas.2313790120
13.
by batterylake 2y ago
The best way to identify the values LLMs hold is not to give them a survey with questions of the form "I believe in the value of justice. Agree or disagree?" Instead you need to present scenarios which put these beliefs into effec
14.
by batterylake 2y ago
I'm interested if this might be a more robust method to resolve conflicting information across retrieved documents. Instead of having the LLM reason over which source to trust, the solution is to incorporate the knowledge into parametr
15.
Mind the Value-Action Gap: Do LLMs Act in Alignment with Their Values?(arxiv.org)
1 points by batterylake 2y ago|1 comments
16.
Parametric Retrieval Augmented Generation(arxiv.org)
3 points by batterylake 2y ago|1 comments
17.
by batterylake 2y ago
This is incredible! PaulPauls, how would you like us to cite your work?