Elamur Research⌕
☰
home / authors / Nils Lukas

Nils Lukas

Nils Lukas has 2 papers in our library, in 2026, most of them on Reasoning; often with Juntian Zhang, Kshitij Mishra, Salem Lahlou.

2 papers in our library · 2026

OpenAlex · ORCID · h-index 6 · 37 works · 394 citations OpenAlex, 2026-10-14

📄 2 works in our library · 🔗 Cited 101 times in our library · 1 more in our library without a page yet · ⭐ Paper Stars ranking: #10967

The profile is the library's match of the name; two people with one name are not always told apart.

Elsewhere: Google Scholar · Semantic Scholar · dblp · ACL Anthology

 Affiliations

🎓Mohamed bin Zayed University of Artificial IntelligenceAbu Dhabi · United Arab Emirates2024–2026 · 5 paperscurrent🎓Fudan UniversityShanghai · China2026 · 3 paperscurrent🎓Harvard UniversityCambridge · United States2026 · 3 paperscurrent🎓Renmin University of ChinaBeijing · China2026 · 3 paperscurrent🎓Massachusetts Institute of TechnologyCambridge · United States2026 · 1 papercurrent🎓Michigan State UniversityEast Lansing · United States2026 · 1 papercurrent🎓University of WaterlooWaterloo · Canada2019–2023 · 13 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine.

 Fields, topics and concepts

Of its works in our library, with the number of works on each.

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 2 worksComputer Vision and Pattern Recognition · 1 workArtificial Intelligence · 1 workArts and Humanities · 1 workLanguage and Linguistics · 1 work

🧭 Topics 1Research subjects and languages the work is about.Reasoning · 2 works

 Works by year

2020s

20262 papers

 Papers

The most cited first (OpenAlex and our library).

  1. Forest Before Trees: Latent Superposition for Efficient Visual Reasoning (2026 · 3 citations)
    Chain-of-Thought helps vision-language models reason step by step. However, spelling out rationales as discrete text loses much of the continuous visual signal. Existing latent-reasoning methods avoid text but still train each hidden state to predict one exact next token, which pushes the model to commit to specific…
  2. SD-E2: Semantic Exploration for Reasoning Under Token Budgets (2026)
    Small language models have a hard time with multi-step reasoning when token budgets are tight. They need to try different solution approaches, but sampling many traces is costly. Standard RL rewards such as GRPO also cannot tell a new strategy apart from a reworded old one. The authors propose SD-E2, an RL framework…

All works

3 works of theirs in our library, the most cited first; 2 with a page here, the others link their original until we summarise them.

  1. Analyzing Leakage of Personally Identifiable Information in Language Models ↗arXiv:2302.00539 · no summary yetNils Lukas et al.2023 · 252 citations
  2. Forest Before Trees: Latent Superposition for Efficient Visual ReasoningAnnual Meeting of the Association for Computational Linguistics (ACL)Yubo Wang, Juntian Zhang, Yichen Wu, et al.2026 · 3 citations
  3. SD-E2: Semantic Exploration for Reasoning Under Token BudgetsFindings of the Association for Computational LinguisticsKshitij Mishra, Nils Lukas, Salem Lahlou2026

 Co-authors here

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

Venues

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

 Cited by

Papers of our library cite their works 101 times (of the 149,439 papers whose bibliographies we read); 31 citing papers with a page here →.

Updated

This page in Markdown → https://md.research.elamur.ai/free/en/authors/nils-lukas/