iiinova — where your digital self keeps evolving

At the domain level, immigration is driven by character portrayals (who is being portrayed and how) while gun control is driven by judicial conflict and institutional enforcement schemas (what is happening), suggesting the two domains are organized around different narrative dimensions. 🧵9/10

We find that for immigration, characters play a central role in driving frame prediction, particularly immigrants. In contrast, gun control frame predictions are driven by narrative clusters, particularly those dealing with judicial and legal schemas.

At the frame level, we found narrative signatures vary meaningfully across policy frames. For example, the Legality/Constitutionality frame in gun control is dominated by judicial conflict schemas, suggesting legal framing emphasizes courtroom battles while political actors remain peripheral. 🧵8/10

Legal framing of gun control coverage tends to emphasize courtroom battles rather than legislative politics.

The SHAP analysis operates at three levels: instance, frame, and domain. At the instance level, narrative schema features capture equivalent information to RoBERTa embeddings, while offering more succinct and interpretable representations. 🧵7/10

SHAP feature importance comparison between RoBERTa embeddings and narrative features for frame prediction on a single test instance, showing both approaches capture similar semantic patterns.

Main findings: 1. Structured clustering outperforms k-means on frame prediction and cluster purity, indicating our schemas carry high predictive signal. 2. Narrative features based on induced schemas match black box models in predictive performance, while offering greater interpretability. 🧵6/10

We evaluated 4,000 news articles across immigration and gun control. Schema quality was validated by a trained linguist, with 94-96% rated high quality, consistently capturing Entman's framing elements. We also conducted a SHAP analysis using induced schemas as features for frame prediction. 🧵5/10

Examples of high-quality generated schemas, evaluated on their overall coherence, as well as on striking a balance between coverage of and specificity to the sentences included in the context.

Textual similarity alone can group narratives that look alike but represent different framings of the same events. Our structured clustering method uses character-role configurations as cannot-link constraints to ensure narratives with conflicting framings are appropriately separated. 🧵4/10

We present a framework for unsupervised, domain-agnostic narrative schema induction that scales to any large corpora. Our method extracts causal event chains, assigns character roles (Hero/Threat/Victim) to entities, and uses these as constraints in a structured clustering framework. 🧵3/10

For a large scale news corpus, we first construct narrative event chains and obtain character and role annotations for them. We then cluster these narrative chains using the character and role information as constraints. The generated narrative clusters are representative of fine-grained and nuanced narrative schemas.

Existing NLP approaches often build narratives bottom-up from extractable atomic units like predicate argument structures or entities. While highly scalable, these methods seldom capture the evaluative and ideological dimensions central to how meaning is constructed in the media. 🧵2/10

Law enforcement spending framed as worker protection versus government waste across policy domains.

Computational approaches to media narrative analysis either miss nuanced storytelling patterns through coarse-grained analysis, or require domain-specific taxonomies that limit scalability. We show joint event and character modeling can address this gap. Details in our #ACL2026 (Main) paper. 🧵1/10

Paper Title: A Structured Clustering Approach for Inducing Media Narratives

Authors: Rohan Das, Advait Deshmukh, Alexandria Leto, Zohar Naaman, I-Ta Lee, Maria Leonor Pacheco

12411

Main Takeaways: 1. Synchronous collaboration produces more consistent and cohesive themes, especially on semantically homogeneous data. 2. Systems with richer user control benefit more from real-time deliberation. 3. Dataset characteristics shape outcomes as much as collaboration setting. 🧵6/7

111

Our evaluation framework operates across 3 dimensions: 1. Consistency: Do coders across modalities surface the same themes? 2. Cohesiveness/Distinctiveness: Are docs within a theme more similar to each other than to docs in other themes? 3. Correctness: Are automated assignments accurate? 🧵4/7

100

Our contributions: 1. A 3-pronged evaluation framework for interactive qualitative coding systems. 2. Experimental results comparing synchronous and asynchronous coding. 🧵3/7

In this study, we measure the quality of coded themes using different interactive systems under different coding configurations.

We tasked researchers to use NLP tools to identify themes from a dataset of 5.5k Facebook ads on climate change. Here are two of the resulting code-books. Do you think one code-book is better than the other? 🧵2/7

Themes as they belong to two different qualitative codebooks derived from the same dataset.

100

Rohan @rohandas.net · 05/05/2026

What’s the best way to analyze online discourse on any given topic? Is there a right way to use NLP tools to sift through massive datasets? To find out, we tested several tools across different collaboration settings and report findings in an #ACL2026 (Main) paper: arxiv.org/abs/2408.09030 🧵1/7

Paper Title: Effects of Collaboration on the Performance of Interactive Theme Discovery Systems

Authors: Alvin Po-Chun Chen, Rohan Das, Dananjay Srinivas, Alexandra Barry, Maksim Seniw, Maria Leonor Pacheco

1125

Main Takeaways: 1. Synchronous collaboration produces more consistent and cohesive themes, especially on semantically homogeneous data 2. Systems with richer user control benefit more from real-time deliberation 3. Dataset characteristics shape outcomes as much as collaboration setting does 🧵4/5

100

Rohan @rohandas.net · 05/05/2026

We introduce an evaluation framework across 4 dimensions: 1. Consistency: Do coders across modalities surface the same themes? 2. Cohesiveness: Are documents within a theme similar? 3. Distinctiveness: Are themes different from each other? 4. Correctness: Are automated assignments accurate? 🧵3/5

100

Rohan @rohandas.net · 05/05/2026

We evaluated 3 NLP-assisted coding systems (topic model, relational, LLM-based) under synchronous vs. asynchronous collaboration, on 2 corpora: 85K COVID vaccine tweets and 5.5K climate change ads. The study involved 33 researchers across 2 universities, spanning 30 coding experiments. 🧵2/5

Boulder NLP @bouldernlp.bsky.social · 03/11/2025

Reposted by Rohan

Strong showing from Boulder NLP at #EMNLP2025! 🚀 Come find us at these sessions to discuss morphology, semantics, LLMs, cultural analytics, HCI, and more. See you in Suzhou! 🇨🇳

0104

Rohan @rohandas.net · 01/11/2025

Also check out Grameen Bank which is a microfinance org that caters to the poor and is owned by the borrowers. Wiki: en.wikipedia.org/wiki/Grameen...

010

Rohan @rohandas.net · 01/11/2025

In the Indian context, Amul stands out as a shining example of the co-operative model that revolutionized the dairy industry. Wiki: en.wikipedia.org/wiki/Amul

Sundance Institute @sundance.org · 27/03/2025

Reposted by Rohan

Big news! Sundance Film Festival reveals Boulder, Colorado as our new location starting in 2027 and beyond. The annual event by the nonprofit Sundance Institute will continue to entertain and inspire audiences through independent film. 🎬 Read more: sndnc.org/boulder2027 #SundanceFilmFestival

129138

Reposted by Rohan

Lauren Klein @laurenfklein.bsky.social · 03/03/2025

Already a time capsule, but back in January a bunch of us working at the intersection of the humanities and AI/ML came together to sketch out eight provocations from the humanities for genAI research. Here's a 🧵 1/ arxiv.org/abs/2502.19190

arxiv.org

Provocations from the Humanities for Generative AI Research

This paper presents a set of provocations for considering the uses, impact, and harms of generative AI from the perspective of humanities researchers. We provide a working definition of humanities res...

511444

Reposted by Rohan

chenhaotan.bsky.social @chenhaotan.bsky.social · 24/01/2025

Spent a great day at Boulder meeting new students and old colleagues. I used to take this view every day. Here are the slides for my talk titled "Alignment Beyond Human Preferences: Use Human Goals to Guide AI towards Complementary AI": chenhaot.com/talks/alignm...

0165

IDE: I can't recommend PyCharm enough for Python heavy workflows. The good folks at JetBrains truly deserve some flowers, and I think everything they build is chef's kiss! I love PyCharm's debugger and python shell implementations (I dislike notebooks with a vengeance.)

010

Terminal Emulator: A switch to Termius last year massively sped up my somewhat complex workflow that interfaces with CU's HPC cluster. It has mostly everything that Warp offers (I think?) + smart SSH and FTP client, port forwarding, and automation routines.

dodgy solution for CU's implementation of O365 Mail/Exchange Online.

Email: Having tried tons of different email clients over the years, I like Spark Mail the best (which is what I use on my phone/tablet, and probably one of the few services I would pay a subscription fee for. But they don't support Linux, and so on my laptop I am stuck with Thunderbird with a

Browser: I switched to Brave last year after having used Firefox forever. It is chromium-based, (supposedly) privacy-centric, and tabs live indefinitely by default. However, I think Vivaldi is the most like for like replacement for Arc - never tried it since it is an overkill for my needs.

Rohan @rohandas.net · 19/11/2024

Bonus: You get to work out of a lab with not only windows but also gorgeous views of the Rockies! ⛰️

010

Rohan @rohandas.net · 19/11/2024

If you're applying to NLP PhD programs, consider Maria’s new group at CU Boulder - she’s amazing, and the larger Boulder NLP community is a great place to learn, grow, and collaborate! Feel free to reach out if you want to chat about the CS PhD program or life in Boulder! My DMs are open.

220

Reposted by Rohan

Joe Stacey @joestacey.bsky.social · 18/11/2024

After going to NAACL, ACL and #EMNLP2024 this year, here are a few tips I’ve picked up about attending #NLP conferences. Would love to hear any other tips if you have them! This proved very popular on another (more evil) social media platform, so sharing here also 🙂 My 10 tips:

148316

Rohan @rohandas.net · 14/11/2024

Happy to report that the event was a disaster. Took us an hour to even get into the venue. Long winding queues everywhere and barely any room to stand. 🤡