TokenVerse: Versatile Multi-concept Personalization in Token Modulation Space
Paper
•
2501.12224
•
Published
•
46
Yonatan Bitton
Yonatan-Bitton
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
about 1 month ago
TokenVerse: Versatile Multi-concept Personalization in Token Modulation
Space
authored
a paper
7 months ago
Visual Riddles: a Commonsense and World Knowledge Challenge for Large
Vision and Language Models
Organizations
Yonatan-Bitton's activity
upvoted
a
paper
about 1 month ago
upvoted
a
paper
7 months ago
authored
a
paper
7 months ago
upvoted
an
article
7 months ago
Article
Docmatix - a huge dataset for Document Visual Question Answering
•
72
upvoted
a
paper
8 months ago
Update app.py
#7 opened 8 months ago
by
roopalgarg
updated
a
Space
8 months ago
Update README.md
#6 opened 10 months ago
by
roopalgarg
Update app.py
#5 opened 10 months ago
by
roopalgarg
Update app.py
#4 opened 10 months ago
by
roopalgarg
update app to support additional IIW data release sets.
#3 opened 10 months ago
by
roopalgarg
updated
a
Space
10 months ago
reacted to
gsarti's
post with 👍
about 1 year ago
Post
🔍 Today's pick in Interpretability & Analysis of LMs: A Chain-of-Thought Is as Strong as Its Weakest Link: A Benchmark for Verifiers of Reasoning Chains by
@alonjacovi
@yonatanbitton
B. Bohnet J. Herzig
@orhonovic
M. Tseng M. Collins
@roeeaharoni
@mega
This work introduces a new methodology for human verification of reasoning chains and adopts it to annotate a dataset of chain-of-thought reasoning chains produced by 3 LMs. The annotated dataset, REVEAL, can be used to benchmark automatic verifiers of reasoning in LMs.
In their analysis, the authors find that LM-produced CoTs generally contain faulty steps, often leading to incorrect automatic verification. In particular, CoT-generating LMs are found to produce non-attributable reasoning steps often, and reasoning verifiers generally struggle to verify logical correctness.
📄 Paper: A Chain-of-Thought Is as Strong as Its Weakest Link: A Benchmark for Verifiers of Reasoning Chains (2402.00559)
🔡 Dataset: google/reveal
This work introduces a new methodology for human verification of reasoning chains and adopts it to annotate a dataset of chain-of-thought reasoning chains produced by 3 LMs. The annotated dataset, REVEAL, can be used to benchmark automatic verifiers of reasoning in LMs.
In their analysis, the authors find that LM-produced CoTs generally contain faulty steps, often leading to incorrect automatic verification. In particular, CoT-generating LMs are found to produce non-attributable reasoning steps often, and reasoning verifiers generally struggle to verify logical correctness.
📄 Paper: A Chain-of-Thought Is as Strong as Its Weakest Link: A Benchmark for Verifiers of Reasoning Chains (2402.00559)
🔡 Dataset: google/reveal
image urls failed to load
1
#2 opened about 1 year ago
by
luodian
Thank you
2
#1 opened over 1 year ago
by
mikehemberger
Upload docci.csv
#1 opened over 1 year ago
by
Yonatan-Bitton