Serra Sinem Tekiroğlu

Senior Data Scientist @twain. Former researcher at the Language Dialogue Technologies (LanD) group at FBK

[Rada Mihalcea @radamihalcea]

Today’s “NLP@Michigan” keynote by the amazing @MonaDiab77 gave me a LOT to think about: the field of NLP is going through an identity crisis. Many of the deeply human layers of meaning associated with language have been conflated by treating language as “data,” and we often end up optimizing for the wrong things — benchmark performance, commercial revenue & other such narrow objectives. But there are ways to address this, and we can still “steer the ship” by recognizing there is more to language than just “data.” ✨

[LanD @LanD_FBK]

Last week marked the final conference of the #Hatedemics project, with Marco Guerini (@m_guerini ) as coordinator and Helena Bonaldi (@helena_bonaldi ). A great opportunity to share our results and celebrate the impact and success of the project 🎉

[LanD @LanD_FBK]

Exciting day at #INLG2025 in Hanoi! 🇻🇳

Daniel Russo (@russ_eau) and Stefano Menini (@stefano_menini) presented "Face the Facts! Evaluating RAG-based Pipelines for Professional Fact-Checking"

🔗 https://t.co/u4P3MGhYwd

[Mohamed Chahin @mo_chahin1]

We're looking for a Founding Growth Lead!

We pivot fast. Making big bets, forcing our way towards PMF.

Here are four facts that make Twain special. Why you should join

1/4 🧵 Raised $8M from @sequoia @TQVentures but kept the team to only 7 FTE (incl. me)

[Serra Sinem Tekiroğlu @serrasinem]

@ahmerivahmeri @ozgeppn Ah keske be!

[Serra Sinem Tekiroğlu @serrasinem]

@ahmerivahmeri @ozgeppn Ah keske be!

[LanD @LanD_FBK]

What happens when Harry meets Superman instead of Sally for a chat? 🧔🦸‍♂️

We explore how LLMs adapt their responses not just to their given persona, but also to their interlocutors’.

Excited to share our #ACL2025NLP (main) paper by @docchipinti8 @m_guerini @MalvinaNissim

1/🧵

[CSHAM Tutorial (ACL2025) @csham_tutorial]

We are happy to announce our tutorial, "NLP for Counterspeech against Hate and Misinformation".

It will be held on July 27, 2025, at #ACL2025NLP in Vienna, Austria!

Stay tuned for more details 👀
https://t.co/C3YGFtlahV

[Marco Guerini @m_guerini]

This year we are organising a tutorial for all those who want to have a better understanding on how counterspeech can be used in #NLProc for fighting both Hate and Misinformation. If you are at #ACL2025NLP and interested join us!

[Serra Sinem Tekiroğlu @serrasinem]

@ahmerivahmeri @mentalolarakk Mis gibi bir yaz aksaminda seninle soyle dolasabilsek, sakince, keyifle @ahmerivahmeri

[Serra Sinem Tekiroğlu @serrasinem]

@ahmerivahmeri @tarihiiimerak Gel gotureyim, Otzi'ye yurume mesafesindeyim😅

[Sara Tonelli @sara_hlt]

My very first time in a technical meeting with female researchers only 💪💪💪#NLProc #womenintech @LanD_FBK @chub_fbk

[Helena Bonaldi @helena_bonaldi]

🚨 Our paper “Is Safer Better? The Impact of Guardrails on the Argumentative Strength of LLMs in Hate Speech Countering” has been accepted at #EMNLP2024!

With @grexit_ @NBenjaminOcampo @ECabrio @serena_villata @m_guerini

https://t.co/KaaYbCCbYw

A thread 🧵

[LanD @LanD_FBK]

#LT2024FBK day. Some pictures of our students from LanD presenting their work. Well done folks!

[MT Group at FBK @fbk_mt]

The Language Technology Research Highlights @ FBK 2024 has just started. Our keynote speaker @barbara_plank is ready to present "Are LLMs narrowing our Horizon? Let's embrace variation in NLP!"

😌👉https://t.co/38IG1HD2ts

#nlproc #LT2024FBK

[LanD @LanD_FBK]

6 papers accepted at #EMNLP2024 submitted by @FBK_research: 2 from our group, plus others from @DH_FBK, @fbk_stek and @fbk_mt… congrats to all! 🥳

[Serra Sinem Tekiroğlu @serrasinem]

@ahmerivahmeri @erelender 🤣🤣

[Serra Sinem Tekiroğlu @serrasinem]

@ahmerivahmeri beni de ikna etti.

[LanD @LanD_FBK]

Kudos to @docchipinti8! During her virtual poster presentation at #NAACL2024 she introduced our PRODIGy dataset and she gave many insights into how profile representations impact dialogue models! For sure this is just the first step of a successful career in #NLProc! 💡

[Serra Sinem Tekiroğlu @serrasinem]

@m_guerini We started together, but you guys nurtured it into this amazing stage

[Serra Sinem Tekiroğlu @serrasinem]

A great work from LanD 💗

[LanD @LanD_FBK]

✏️ What happens if we apply human correction to synthetic data for fine-tuning LLMs?

We are happy to share that our paper "Fine-tuning with HED-IT: The impact of human post-editing for dialogical language models" was accepted at #ACL2024! 🎉🚀

https://t.co/hOcdz89FUx

🧵[1/7]