The PhotoBook Dataset: Building Common Ground through Visually-Grounded Dialogue (2019 · 53 citations)
Current visually grounded dialogue systems tend to lose consistency across a conversation, partly because they do not exploit the shared history that human partners build up.
Word Representation Learning in Multimodal Pre-Trained Transformers: An Intrinsic Evaluation (2021 · 25 citations)
Pre-trained vision-and-language Transformers are said to learn general-purpose word meanings, but it was unclear how well these meanings match human intuitions.
Common Objects Out of Context (COOCo): Investigating Multimodal Context and Semantic Scene Violations in Referential Communication (2026)
The paper asks how much vision-language models (VLMs) draw on surrounding scene context when naming an object, and whether this depends on how well the object fits the scene.

All works

4 works of theirs in our library, the most cited first; 3 with a page here, the others link their original until we summarise them.

The PhotoBook Dataset: Building Common Ground through Visually-Grounded DialogueAnnual Meeting of the Association for Computational Linguistics (ACL)Janosch Haber, Tim Baumgärtner, Ece Takmaz, et al.2019 · 53 citations
Word Representation Learning in Multimodal Pre-Trained Transformers: An Intrinsic EvaluationTransactions of the Association for Computational Linguistics (TACL)Sandro Pezzelle, Ece Takmaz, Raquel Fernández2021 · 25 citations
Common Objects Out of Context (COOCo): Investigating Multimodal Context and Semantic Scene Violations in Referential CommunicationTransactions of the Association for Computational Linguistics (TACL)Filippo Merlo, Ece Takmaz, Wenkai Chen, et al.2026
When Context Misleads: Surprisal, Energy and Attention Entropy as Metrics of Coherence Illusions in LLMs ↗arXiv:2606.21203 · no summary yetEce Takmaz et al.2026

Co-authors here

5 co-authors with a page, the most papers together first.

Where their works here were published, the most works first, with the editions of those that have a page.

Papers of our library cite their works 194 times (of the 149,439 papers whose bibliographies we read); 52 citing papers with a page here →.

Updated 1 November 2026