Nir Grinberg

🚨New paper🚨out in @socialmedia-soc.bsky.social w/ John Wagner and @jenny-oser.bsky.social 🎉 We all know who's a political influencer on social media, right? Well... we may THINK we do, but what we're actually measuring is very different. How different? 🧵👇

View full thread

The overarching story here is quite simple: To build generalizable knowledge about social media influencers, we need clearer, multidimensional frameworks (rather than relying on single-metric measures), and we need to be more transparent about what we actually measure.

journals.sagepub.comHow Definitions Shape Our Understanding of Political Influencers on Social Media - Jennifer Oser, John Wagner, Nir Grinberg, 2026Political discourse now circulates through sprawling social media networks where multiple actors wield influence in addition to traditional elites. Yet scholarl...

Read the full open-access article here: journals.sagepub.com/doi/10.1177/.... We are grateful for comments received from participants of the Digital Democracy Workshop at UZH, MPSA, and ICA conferences, and ERC funding.

The dimensions of influence are not interchangeable and have non-trivial relationships. E.g. Engagement selects those posting more frequently, but Elite status does not.

Importantly, upstream definitional choices fundamentally alter the downstream audience demographics. "Network Importance" influencers reach significantly older, whiter audiences, while Follower and engagement-based definitions yield significantly younger audiences.

Definitions vary wildly in size – from a few hundred political influencers to 100,000 or more – and even within the same order of magnitude, they rarely capture the same political influencers (often <40% overlap).

Definitions also determine how POLITICAL political influencers really are. Elite & centrality-based defs focus on the more politically vocal, while follower & engagement-based defs capture more of the long tail.

🚨 Excited to share our new paper to appear in the Findings of ACL 2026! "JudgeMeNot" asks: Can LLMs learn the specific reasoning signature of an individual judge? We tackle this in the low-resource setting of Hebrew case law. 🔗 arxiv.org/abs/2604.18041 🧵👇

7/8 🛤️ An important caveat is that our work successfully emulates sentence-level, granular reasoning, which is considerable easier that modeling the full trajectory of case.

arxiv.orgJudgeMeNot: Personalizing Large Language Models to Emulate Judicial Reasoning in HebrewDespite significant advances in large language models, personalizing them for individual decision-makers remains an open problem. Here, we introduce a synthetic-organic supervision pipeline that trans...

8/8 🎉 Super proud of this computational social science work out of our lab at @bengurionuniv.bsky.social ! Huge thanks to my co-authors Itay Razumenko and Arnon Sturm. Read the full paper at arxiv.org/abs/2604.18041 If you're attending ACL, come say hi 👋 #NLProc #ACL2026 #LegalTech

5/8 📊 The results? CoLA consistently outperforms all baselines. RAG captures surface-level style (like POS distributions), but fails in reasoning. CoLA-generated reasoning is so faithful that an LLM trained specifically on each judge cannot tell the difference from the actual judge. 🤯

6/8 While all data analyzed here is entirely public, by public officials operating in their official capacity for public scrutiny, the sensitivity of the domain + personalization requires further ethical considerations beyond simple regulatory compliance as detailed in Section 9 of the paper.

3/8 🛠️ The bottleneck for personalization is the lack of reasoning supervision. To fix this, we built a "synthetic-organic" pipeline. We use an agentic workflow to extract raw reasoning sentences from unstructured Israeli verdicts and generate targeted synthetic questions that prompted them.

Ben-Gurion University of the Negev

4/8 🧠 For training, we used Chain-of-LoRA (CoLA) in two steps: 1️⃣ Causal LM tuning (adapts to general writing style). 2️⃣ Instruction tuning (focuses on legal reasoning). All using a single GPU and Gemma-4B model, which was super fast.

2/8🧑‍⚖️ Judges aren't interchangeable; they have distinct judicial philosophies. Yet, most Legal NLP treats them as one generic voice. The result? AI "mode collapse." Standard models generate long, encyclopedic text that completely misses the human legal fingerprint.

Nir Grinberg reposted

@bengurionuniv.bsky.social

9 Jan

Can social media detect economic shocks before official data does? A new PNAS Nexus study led by @nirg.bsky.social and Samuel Fraiberger shows that AI models tracking job-loss disclosures on social media can predict U.S. unemployment insurance claims up to two weeks early,

doi.orgCan social media reliably estimate unemployment?Abstract. Digital trace data hold tremendous potential for measuring policy-relevant outcomes in real-time, yet its reliability is often questioned. Here,

New paper out in @pnasnexus.org: We show how skewed social media data can still be used to reliably estimate unemployment, not just nationally but down to the city level. 📈 doi.org/10.1093/pnas... 1/N

Kudos to my wonderful co-authors Do Lee linkedin.com/in/do-lee and @manueltonneau.bsky.social (both on the job market!), Boris Sobol il.linkedin.com/in/boris-sobol and Sam Fraiberger samuelfraiberger.com. 11/N

Credit is also due to @davidlazer.bsky.social for prompting Sam & I to think about this problem 7(!) years ago ;) 12/fin

There is clear public value here, potentially extending to other countries, especially where official statistical systems are under-developed. 9/N

Yet platform data-access policies increasingly block this potential. Whether platforms or regulators will enable change in the coming years is a core policy question. 10/N

Key finding 3: This also works at the state and city (!) level, including "holdout cities" where official UI numbers are sparse or irregularly updated. As expected, accuracy scales with platform penetration and unemployment shocks. 7/N

Why this matters? Beyond forecasting, this approach can provide early warnings, surface local labor market stress hidden by national averages, and help flag measurement issues in real time. 8/N

Key finding 1: Capturing linguistic diversity matters. Training LLMs with active learning lets us detect many more ways people talk about job loss, producing a far more representative sample of unemployed users than existing approaches. 5/N

Key finding 2: Our approach consistently outperforms industry consensus forecasts and can improve predictions of US UI claims up to two weeks ahead of official releases. That’s two weeks of additional lead time for policymakers. 6/N

So we ask a hard question economic actors and policymakers rightly worry about: Can skewed social media data be turned into trustworthy indicators of unemployment? Can we produce robust predictions across geography ✅, time ✅, demography ✅, and forecasting horizon ✅ ? 3/N

aclanthology.orgMultilingual Detection of Personal Employment Status on TwitterManuel Tonneau, Dhaval Adjodah, Joao Palotti, Nir Grinberg, Samuel Fraiberger. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

We combine JoblessBERT (an encoder LLM developed in previous work aclanthology.org/2022.acl-lon... which detects ~3× more employment-related content without sacrificing precision) with post-stratification using inferred demographics to correct for platform bias. 4/N

Why this matters: In March 2020, weekly unemployment insurance claims jumped from 278K to nearly 6 million in two weeks. As official data lagged, policymakers were flying blind about where the shock was hitting and who was being affected. 2/N

Nir Grinberg reposted

Kevin Yang

@yang3kc.bsky.social

17 Jan 2025

Introducing “DomainDemo: a dataset of domain-sharing activities among different demographic groups on Twitter.” Today, we release five derived metrics of over 129,000 domains, quantifying their characteristics such as geographical reach and audience partisanship. 1/3

Nir Grinberg reposted

Ross Dahlke

@rossdahlke.bsky.social

17 Jan 2024

Incels (involuntarily celibates) are increasingly using violent language, particularly non-directed violent language in the largest incel forum, finds @danielmatter.bsky.social @miriamschirmer.bsky.social @nirg.bsky.social @jurgenpfeffer.bsky.social arxiv.org/abs/2401.02001

🚨New paper🚨 out in the International Journal of Press/Politics w/ Assaf Shamir and @jenny-oser.bsky.social 🎉 Here's what we learned from studying the composition of political content available to 600k+ registered U.S. voters on Twitter during the 2020 election. doi.org/10.1177/1940... 🧵👇

Big shout-out to the people whose work enabled this research, including @sdmccabe.com @jongreen.bsky.social @davidlazer.bsky.social Magdalena Wojcieszak @jatucker.bsky.social Subhayan Mukerjee @ylelkes.bsky.social @kthorson.bsky.social @chriswells.bsky.social (pls tag others if missing). 6/

We are also grateful for comments received on earlier versions of this work from Diyi Liu, Eran Amsalem @patyrossini.bsky.social Alon Zoizner, and @orentsur.bsky.social & for funding from European Research Council (ERC), Israel Science Foundation (ISF) and BGU's Data Science Center.

Even when putting aside the more extreme "media superconsumers", the two media-oriented clusters (which are ~20% of the population), get half or more of their political content *directly* from media organizations and journalists, without any mediation from peers. 4/

Finally, looking at the demographic composition of consumption "types", we find that the media-oriented clusters (exc. superconsumers) have older individuals, more women, and more registered Democrats. 5/

People's political feeds mostly map onto 8 distinct types that vary in the amount of politics they get, both in absolute #'s and as % the feed as a whole. Still, for nearly 90% of the population, about 1 in 12 posts from their network are political. Quite an engaged public! 2/