Niccolò Campolungo

Decoding the future @ Litus AI ~ NLP PhD @ Sapienza - Engineer at heart - he/him.

[Niccolò Campolungo @Valahaar]

Super happy that our DiBiMT paper got recognized as the Best Resource Paper @aclmeeting #ACL2022! 🥳🎉

A (not so) short 🧵 about it :) 👇
Joint work with @FedeMartelli25 @FrancescoSaina @RNavigli

📜: https://t.co/MJI13hgOID

1/7

[Niccolò Campolungo @Valahaar]

@peternixey @msfeldstein @cursor_ai Could I have that zip as well? Looks super interesting :)

[Valentino Maiorca @ValeMaiorca]

📢 It looks like relative representations are here to stay!

I'm beyond thrilled to announce that our work has been selected as one of the notable top 5% (oral) papers at #iclr23 ! 🥳

https://t.co/nlZBiaIMHZ

[1/5]

[SapienzaNLP @SapienzaNLP]

We reframe #MT evaluation as sequence tagging in our #WMT2022 paper: check MaTESe, an interpretable metric that tags error spans in translations!

📜 https://t.co/e8nBm1c2Ow

By @19Stefano97 @LorenzoProiet13 @alescire94 @Valahaar @RNavigli @babelscape @ERC_Research #EMNLP2022

[Pere-Lluís Huguet Cabot @PereLluisHC]

Great to wake up on a lazy Sunday and see REBEL is still doing numbers. This is the page for top seq2seq models at @huggingface, currently top 12 with 200K monthly downloads. Thank you to whomever is using it and honestly curious to know what people are using it for.

[SapienzaNLP @SapienzaNLP]

Interested in reducing semantic bias in NMT models with smart data creation & the KL divergence?

Come check out the work by @Valahaar, @pasini_t, @emelin_denis & @RNavigli!

Presentation today @ 9.15 PDT / 18.15 CET, session 8D (MT 3) #NAACL2022

📝 https://t.co/XNKPAukqU9

[Leon Derczynski ⚒️☁️🏔️🌲 @LeonDerczynski]

machine learning researchers learn to optimise their own best paper rate through collusion and other unregulated mechanisms

https://t.co/GsZxab8jCG

[Riccardo Orlando @RiccardoRicOrl]

Hey #NLProc, I built this little tool to make working with @huggingface 🤗Transformers a bit easier. If you want to directly access whole-word embeddings hassle-free, give it a try!

👉GitHub: https://t.co/HHuR0KKsY9

[Roberto Navigli @RNavigli]

The Rome Workshop on 10 Years of #BabelNet & Multilingual Neurosymbolic Natural Language Understanding was a great success, with productive in-person discussions, amazing talks & >100 online participants! Thanks!
@ERC_Research @Babelscape @SapienzaNLP @SapienzaRoma @WikiResearch

[Babelscape @babelscape]

🎉10 Years of #BabelNet Workshop is now over:
what a great experience!🎉
Thanks to everyone who participated both in person and online!

@SapienzaNLP @babelscape @ERC_Research
@AccademiaCrusca @RNavigli
#workshop @WikiResearch #BabelNet10anniversary

[Niccolò Campolungo @Valahaar]

I have never clapped my hands as much as I have at the @aclmeeting 2022 Closing Remaks 😂 thank you Bernardo! #ACL2022

[Iacobacci et al. @iiacobacNLP]

Bravo ragazzi

[Niccolò Campolungo @Valahaar]

Presenting in front of a live audience was an amazing experience :) hope we nailed it! Thanks @aclmeeting

[ACL 2026 @aclmeeting]

#acl2022 nlp Best Resource Paper

DiBiMT: A Novel Benchmark for Measuring Word Sense Disambiguation Biases in Machine Translation

(Niccolò Campolungo, Federico Martelli, Francesco Saina and Roberto Navigli)

#acl2022 #NLProc

[Niccolò Campolungo @Valahaar]

@rikvannoord @aclmeeting This is DiBiMT, in case you're interested! 😄

📜 https://t.co/MJI13hgOID

🧵 https://t.co/uD6IvgRBI0

[Niccolò Campolungo @Valahaar]

Super happy that our DiBiMT paper got recognized as the Best Resource Paper @aclmeeting #ACL2022! 🥳🎉

A (not so) short 🧵 about it :) 👇
Joint work with @FedeMartelli25 @FrancescoSaina @RNavigli

📜: https://t.co/MJI13hgOID

1/7

[SapienzaNLP @SapienzaNLP]

🥳 We are proud to share the news that our DiBiMT paper on Disambiguation Biases in MT received the ✨Best Resource Paper Award✨ @aclmeeting #ACL2022! See you there!

[Niccolò Campolungo @Valahaar]

We will present DiBiMT on the 25th, during the Best Paper Oral Session at 14.45 Dublin time!

Or, you can come have a chat during Poster Session 6: Resources and Evaluation on the same day, from 10:45 to 12:15! Come say hi 👋😄

7/7

[Niccolò Campolungo @Valahaar]

Super happy that our DiBiMT paper got recognized as the Best Resource Paper @aclmeeting #ACL2022! 🥳🎉

A (not so) short 🧵 about it :) 👇
Joint work with @FedeMartelli25 @FrancescoSaina @RNavigli

📜: https://t.co/MJI13hgOID

1/7

[Niccolò Campolungo @Valahaar]

- (neural) models are pretty confident of their incorrect translations: >90% of the time, they deem their BAD❌ translation better than a GOOD ✅ one!
- In neural Transformer models, disambiguation does not seem to be happening on the encoder side 🤔

6/7