Daniel Cohen retweeted
I will be presenting this paper at SIGIR Virtual Short Paper Session 1 tonight (3:30 AM EDT/12:30 PST)
welcome to attend and discuss Constrained Generation and its usage in web search🔍
I am also looking for research internship roles for Summer 2024🙂 hope to connect !!!
happy to share our📜A Lightweight Constrained Generation Alternative for Query-focused Summarization - has been accepted to @SIGIR2023 as a short paper. work w/ @dcohenIR and more details 👇🧵
Excited to share our SIGIR preprint on leveraging a low-complexity way of controlling the generation of LLM for snippet generation! We achieve close to SoTA performance while using only a saliency map and beam search modification: arxiv.org/pdf/2304.11721.pdf
Draft made available on arXiv ! 👇arxiv.org/pdf/2304.11721.pdf
We picture this constrained generation approach as a way to address the classical snippet problem in web search 🔍 and to help reduce users' cognitive overhead in search sessions 📃.
Daniel Cohen retweeted
happy to share our📜A Lightweight Constrained Generation Alternative for Query-focused Summarization - has been accepted to @SIGIR2023 as a short paper. work w/ @dcohenIR and more details 👇🧵
Daniel Cohen retweeted
[Short Paper] "Inconsistent Ranking Assumptions in Medical Search and Their Downstream Consequences" by @dcohenIR, Du, Mitra, Mercurio, Rekabsaz, and @CarstenEickhoff. This is an early exploration into how ranking model uncertainty may disparately impact under-represented groups.
Daniel Cohen retweeted
Monday, 11th October, 15:00 pm (GMT) Daniel Cohen
@dcohenIR, from Brown University, will be giving an #IRTalk @GlasgowCS @ir_glasgow on "Efficient Uncertainty and Calibration Modeling for Deep Retrieval Models". Event details at samoa.dcs.gla.ac.uk/events/v…. Registration opened.
The #sigir2021 preprint is out! arxiv.org/pdf/2105.04651.pdf
Excited to share that our full paper "Not All Relevance Scores are Equal: Efficient Uncertainty and Calibration Modeling for Deep Retrieval Models" was accepted at #sigir2021!
Coauthors: @UnderdogGeek Oleg Lesota @navidrekabsaz @CarstenEickhoff.
Stay tuned for the preprint!
Daniel Cohen retweeted
I wrote a blog post that reports and summarizes some statistics on the #SIGIR2021 accepted papers. Check it out if you're interested: hamedz.ir/blog/2021/sigir21-…
@SIGIRConf @SIGIRForum @sigir_students @ACMSIGIR
A thread 👇
I'm also happy to share that my first single author paper, "Allowing for The Grounded Use of Temporal Difference Learning in Large Ranking Models via Sub State Updates" was also accepted as a full paper at #sigir2021!
27
Excited to share that our full paper "Not All Relevance Scores are Equal: Efficient Uncertainty and Calibration Modeling for Deep Retrieval Models" was accepted at #sigir2021!
Coauthors: @UnderdogGeek Oleg Lesota @navidrekabsaz @CarstenEickhoff.
Stay tuned for the preprint!
Daniel Cohen retweeted
Chinese researchers (and one from Australia) put together a dataset of Uyghur, Tibetan, and Korean — all ethnicities mainland is oppressing right now. #MachineLearning community should be enraged about this, but instead this journal tags it as a “focus article”. (H/t @slashML)
Daniel Cohen retweeted
#UMassCICS Center for Intelligent Information Retrieval researchers, prof. Bruce Croft (@wbc11) and PhD students Daniel Cohen and Hamed Zamani, have received a competitive @TechAtBloombergData Science Research Grant Program Award. Way to go! https://nitter.cf/t.co/WAb1oqLwE3
Daniel Cohen retweeted
Very detailed notes of #ICML2018 conference. 55 pages from David Abel. david-abel.github.io/blog/po…
Daniel Cohen retweeted
Our paper received the #sigir2018 best short paper reward! Fantastic work by our former intern @dcohenIR \w @UnderdogGeek @wbc11
Daniel Cohen retweeted
Replying to @MSFTResearch
@MSFTResearch blog features some of my latest collaboration with Bing team and more - exciting work on adversarial models for search \w Dan Cohen @UnderdogGeek @wbc11 nitter.cf/msftresearch/status/10…
Microsoft researchers have been exploring adversarial and reinforcement learning-based approaches to information retrieval. See what they're presenting at #SIGIR2018 about the recent work: aka.ms/AA1symr
Daniel Cohen retweeted
Please retweet/like/comment on this call for science without borders or other barriers! #OpenScience #TravelBan nitter.cf/BenCarterette/status/1…
Daniel Cohen retweeted
Check out the list of papers accepted to the #LND4IR workshop:
lnd4ir.github.io/
Congrats to the authors of accepted papers!
#SIGIR2018 @sigir2018
Daniel Cohen retweeted
After a 39-year-career at UMass Amherst, Distinguished Professor Bruce Croft (@wbc11) is setting aside his faculty responsibilities to focus on research and directing the Center for Intelligent Information Retrieval. Congratulations, Bruce! https://nitter.cf/t.co/iAoix1OAS4
Daniel Cohen retweeted
We've shaped an awesome PC for the #LND4IR workshop @sigir2018 (visit our website lnd4ir.github.io/). #LND4IR is non-archival and the submission due is in 1 week (May 4th). We're waiting for your great work. #SIGIR2018
cc @m__dehghani @diazf_acm @dr_hang_li @nick_craswell nitter.cf/HamedZamani/status/971…
Happy to announce: we'll organize a workshop on "Learning from Limited / Noisy Data for Information Retrieval" @sigir2018 #SIGIR2018
w/ @m__dehghani @diazf_acm @dr_hang_li @nick_craswell
Daniel Cohen retweeted
"Analyzing and Characterizing User Intent in Information-seeking Conversations" accepted as a short paper @sigir2018 w/ @liuyangumass @wbc11 @JTrippas @EvisonZhang @qiumh #SIGIR2018






















