Tessera
Folgen

Chris Potts

@cgpotts.bsky.social

Stanford Professor of Linguistics and, by courtesy, of Computer Science, and member of @stanfordnlp.bsky.social and The Stanford AI Lab. He/Him/His. https://web.stanford.edu/~cgpotts/

30 Posts879 Follower307 Folgt
Chris Potts@cgpotts.bsky.social · 2y ago

I am confident OpenAI will become profitable. They are smart, creative, highly incentivized, and well-funded. On the other hand, any app/company that depends on capturing most of the value from OpenAI's models has an uncertain future, like the Twitter apps of old.

Betsy Sneller@betsysneller.bsky.social · 2y ago

Bill Labov died this morning. I'm not coherent enough to talk about how important and influential and brilliant he was. I am very sad. I was so lucky to know him, and I am grateful every day that he (and Gillian, and Walt, etc) built an academic field where kindness is expected.

Stanford NLP Group@stanfordnlp.bsky.social · 2y ago

ReFT: Representation Finetuning for Language Models Spotlight Poster Zhengxuan Wu · Aryaman Arora · Zheng Wang · Atticus Geiger · Dan Jurafsky · Christopher D Manning · Christopher Potts Fri 13 Dec 07:00 PM UTC [West Ballroom A-D]

Stanford NLP Group@stanfordnlp.bsky.social · 2y ago

Papers (partly) from @stanfordnlp at #NeurIPS 2024: Oral: Embodied Agent Interface: Benchmarking LLMs for Embodied Decision Making Manling Li · Shiyu Zhao · Qineng Wang · Kangrui Wang · … · Weiyu Liu · Percy Liang · Li Fei-Fei · Jiayuan Mao · Jiajun Wu Wed 11 Dec 11:50 PM UTC [East Ballroom A, B]

Chris Potts@cgpotts.bsky.social · 2y ago

My primary role as a Department Chair at Stanford has become complaining about bureaucratic overreach at Stanford. I have send dozens of messages on this topic just this quarter. And yet I have still not mastered the spelling of "bureaucratic".

Stanford NLP Group@stanfordnlp.bsky.social · 2y ago

Natural Language Processing—artificial intelligence that uses human language—has been on a roll lately. You’ve probably noticed! So the Stanford NLP Group has been growing, and diversifying into lots of new topics, including agents, language model programs, and socially aware #NLP. nlp.stanford.edu

Group picture of people in the Stanford NLP Group gathered in front of the shores of Lake Tahoe.
Chris Potts@cgpotts.bsky.social · 2y ago

Lena is a masterful Borgesian fiction imagining the first human brain to be captured on disk. The program enters a state of "terror and extreme panic" on boot-up. If you deny that a machine could be sentient, do you deny the story's premise or the potential reality of such terror? qntm.org/mmacevedo

Lena

You can now buy this story as part of my collection, Valuable Humans in Transit and Other Stories. This collection also includes a sequel story, titled "Driver". Russian translation French transla...

qntm.org

Chris Potts@cgpotts.bsky.social · 2y ago

I've been going through the Wordle archives seeking to achieve a mind-meld with WordleBot, and I finally did it (in a puzzle from winter 2024). I appreciate that it even seems to have known this was my goal.

My solution to a Wordle puzzle alongside the solution from WordleBot. Both of us guessed CRANE / SCALE / PLACE (luck 76). The WordleBot says, "Sensational. We are as one."
Chris Potts@cgpotts.bsky.social · 2y ago

For PhD recommendation systems: this year, as in every year in my experience, MIT EECS gets my highest recommendation: (1) one subjective multiple choice question that does not really try to hide its subjectivity behind made-up numbers and (2) letter upload. I wish my own institution's were as good.

Kanishka Misra@kanishka.bsky.social · 2y ago

There's a known bug in how we compute "word" probabilities with subword-based LMs that mark beginnings of words -- as pointed out by Byung-doh Oh and Will Schuler, & @tpimentel.bsky.social and Clara Meister I'm pleased to announce that minicons now includes a fix which runs batch-wise!

Code: from minicons import scorer lm = scorer.IncrementalLMScorer("gpt2-xl", "cuda:0") stimuli = ["I was a matron in France", "I was a mat in France"] # old way, no correction # P.S. gpt2 does not automatically add a bos token at the beginning... lm.token_score(stimuli, bos_token=True, surprisal=True, base_two=True, bow_correction=False) '''Rounded Output [[('<|endoftext|>', 0.0), ('I', 5.85), ('was', 4.28), ('a', 4.67), ('mat', 16.34), ('ron', 1.74), ('in', 2.12), ('France', 11.43)], [('<|endoftext|>', 0.0), ('I', 5.85), ('was', 4.28), ('a', 4.67), ('mat', 16.34), ('in', 10.78), ('France', 10.71)]] ''' # the new way! notice the surprisal of "mat" in both cases lm.token_score(stimuli, bos_token=True, surprisal=True, base_two=True, bow_correction=True) '''Rounded Output [[('<|endoftext|>', 0.0), ('I', 6.30), ('was', 3.84), ('a', 4.68), ('mat', 16.34), ('ron', 2.11), ('in', 1.75), ('France', 11.42)], [('<|endoftext|>', 0.0), ('I', 6.30), ('was', 3.84), ('a', 4.68), ('mat', 21.34), ('in', 5.80), ('France', 10.69)]] '''Screenshot from Oh and Schuler showing surprisal values for the partial sentences "I was a matron in" and "I was a mat in" using GPT-2 XL with leading whitespaces and trailing whitespaces.
Elisa Kreiss@elisakreiss.bsky.social · 2y ago

I'm excited to kick off my Bluesky presence with wonderful news: Our paper "Reference-Based Metrics Are Biased Against Blind and Low-Vision Users' Image Description Preferences" won a Best Paper Award at the NLP for Positive Impact Workshop at EMNLP! Read it here: aclanthology.org/2024.nlp4pi-...

Reference-Based Metrics Are Biased Against Blind and Low-Vision Users’ Image Description Preferences

Rhea Kapur, Elisa Kreiss. Proceedings of the Third Workshop on NLP for Positive Impact. 2024.

aclanthology.org