Paola Cascante-Bonilla
@pcascanteb.bsky.social
840 followers 680 following 13 posts
UMIACS Postdoc - Incoming Assistant Professor @ SBU CS. Vision and Language. Prev. @RiceCompSci @vislang @merl_news @MITIBMLab. EECS Rising Star'23. Former drummer. 🇨🇷 https://paolacascante.com/
Posts Media Videos Starter Packs
Reposted by Paola Cascante-Bonilla
iccv.bsky.social
#ICCV2025 is deeply committed to promoting diversity, equity, and inclusion within our community. As part of this commitment, travel support is available to help broaden participation.

Applications will be reviewed on a rolling basis until August 20, 2025 (anywhere on Earth).
Reposted by Paola Cascante-Bonilla
lingjunz.bsky.social
We introduce a super simple yet effective strategy to improve video-language alignment (+18%): add hallucination correction in your training objective👌
Excited to share our accepted paper at ACL: Can Hallucination Correction Improve Video-language Alignment?
Link: arxiv.org/abs/2502.15079
pcascanteb.bsky.social
A huge shout-out to all the amazing collaborators! @houyu0930.bsky.social, @tristacao.bsky.social, @haldaume3.bsky.social, Rachel Rudinger.

Won't see you in Singapore due to visa issues tho 💁‍♀️
pcascanteb.bsky.social
NLI Improves Compositionality in Vision-Language Models is accepted to #ICLR2025!

CECE enables interpretability and achieves significant improvements in hard compositional benchmarks without fine-tuning (e.g., Winoground, EqBen) and alignment (e.g., DrawBench, EditBench). + info: cece-vlm.github.io
pcascanteb.bsky.social
- Some happiness among chaos: all CVPR reviews are in! 🎉
- Some bitterness among joy: looking forward to the **highly irresponsible reviewers** desk rejections.

Very happy with most reviewers tho (responsive and eager to fix short/uninformative comments)

Now, back to reviewing PhD applications 🥲
pcascanteb.bsky.social
a bit better, but axolotls still win...
Reposted by Paola Cascante-Bonilla
haldaume3.bsky.social
starter pack for the Computational Linguistics and Information Processing group at the University of Maryland - get all your NLP and data science here!

go.bsky.app/V9qWjEi
Reposted by Paola Cascante-Bonilla
hildekuehne.bsky.social
Job alert! 🚨 (a bit special but anyway)

If you are (or know someone who is) a Phd about to graduate or just graduated, and have to skip some in between time, I currently have a PostDoc position that could run for 6 months.

Needed: Experience with publishing at Top A conferences.
Just ping me. 👍
pcascanteb.bsky.social
Oh look, the author is here! Make sure to follow @navitagoyal.bsky.social — amazing work guaranteed! 🦋
pcascanteb.bsky.social
I've been iterating since Sunday.

It's taking a fair amount of time, but I suspect my future self will be grateful.
pcascanteb.bsky.social
This is my first time serving as an AC for a big conference.

Just read this great work by Goyal et al. arxiv.org/abs/2411.11437

I'm optimizing for high coverage and low redundancy—assigning reviewers based on relevant topics or affinity scores alone feels off. Seniority and diversity matter!
Causal Effect of Group Diversity on Redundancy and Coverage in Peer-Reviewing
A large host of scientific journals and conferences solicit peer reviews from multiple reviewers for the same submission, aiming to gather a broader range of perspectives and mitigate individual biase...
arxiv.org
pcascanteb.bsky.social
Very nice view from my office! :)
Reposted by Paola Cascante-Bonilla
vicenteor.bsky.social
Common software development advice that says we should write tests before implementations also applies to LLM-generated code.

PropTest: Automatic Property Testing for Improved Visual Programming.
arxiv.org/pdf/2403.16921
Reposted by Paola Cascante-Bonilla
pcascanteb.bsky.social
Nice day to engage in ICLR discussions! 🦋
pcascanteb.bsky.social
For me, practicing some new drum-hand techniques during short breaks significantly improves my reading and writing concentration span.

I'm currently trying to learn the traditional grip push/pull technique. Any tips or tricks are more than welcome! :)