PeopleofSuperintelligence

Researcher

Pascal Bergsträßer

Pascal Bergsträßer is a theoretical computer scientist whose paper "Transformers are Inherently Succinct," written with Ryan Cotterell and Anthony Widjaja Lin, was named an ICLR 2026 Outstanding Paper. His doctoral work in Kaiserslautern, supervised by Anthony Widjaja Lin and Georg Zetzsche, centered on Ramsey quantifiers and their applications to verification. He completed his PhD in 2025, after papers on Ramsey quantifiers in 2022 and 2023 and a summer 2024 visit to the Simons Institute for the Theory of Computing as a visiting graduate student. In October 2025, he and his coauthors showed that fixed-precision transformers can be exponentially more succinct than linear temporal logic and recurrent neural networks, which makes basic verification problems for transformers EXPSPACE-complete. He is a postdoctoral researcher in the Department of Computer Science at RPTU in Kaiserslautern, working on logic, automata theory and formal verification.

Timeline

  1. May 2022

    Ramsey quantifiers over automatic structures

    In May 2022, he posts a paper on Ramsey quantifiers over automatic structures with Moses Ganardi, Anthony W. Lin and Georg Zetzsche. It studies quantifiers that express the existence of an infinite directed clique, and it links them to reachability with Büchi conditions in regular model checking and to monadic decomposability of automatic relations.

  2. November 2023

    Ramsey quantifiers in linear arithmetic

    In November 2023, he posts "Ramsey Quantifiers in Linear Arithmetics" with Moses Ganardi, Anthony W. Lin and Georg Zetzsche. The paper gives an algorithm for eliminating Ramsey quantifiers from linear integer, linear real and linear integer-real arithmetic, with applications to program termination and to monadic decomposability of SMT formulas.

  3. 2024

    Visiting the Simons Institute

    In summer 2024, he is a visiting graduate student at the Simons Institute for the Theory of Computing, in the Extended Reunion program on theoretical foundations of computer systems. A PhD student in theoretical computer science at the University of Kaiserslautern-Landau, supervised by Anthony W. Lin and Georg Zetzsche, he works on Ramsey quantifiers in first-order theories and their applications to verification, such as monadic decomposability and liveness.

  4. 2025

    A doctorate in Kaiserslautern

    He completes his PhD in 2025 in the group of Anthony W. Lin, professor of theoretical computer science (automated reasoning) at the University of Kaiserslautern-Landau.

  5. October 2025

    Succinctness of transformers

    In October 2025, he posts "Transformers are Inherently Succinct" on arXiv with Ryan Cotterell and Anthony W. Lin. The paper proves that fixed-precision transformers can be exponentially more succinct than linear temporal logic and recurrent neural networks, and doubly exponentially more succinct than finite automata. As a consequence, basic verification problems for transformers, such as emptiness and equivalence, are EXPSPACE-complete.

  6. November 2025

    The REAL tool for Ramsey quantifiers

    In November 2025, he coauthors "Fast Ramsey Quantifier Elimination in LIRA" with Kilian Lichtner, Moses Ganardi, Anthony W. Lin and Georg Zetzsche. The paper presents REAL, a tool that eliminates Ramsey quantifiers in existential linear arithmetic over integers, reals and the mixed case, and extends the FASTer verification tool to liveness checking.

  7. April 2026

    An ICLR 2026 Outstanding Paper

    In April 2026, ICLR names "Transformers are Inherently Succinct," by Pascal Bergsträßer, Ryan Cotterell and Anthony Widjaja Lin, one of its two Outstanding Papers for 2026. The ICLR 2026 Outstanding Paper Committee describes it as a new perspective on the power of the Transformer architecture, based on how succinctly transformers can encode some concepts compared to models such as RNNs.

  8. As of October 2026

    Postdoc in Kaiserslautern

    He is a postdoctoral researcher in the Department of Computer Science at RPTU in Kaiserslautern, with research interests in logic, automata theory and formal verification. He works in the research group of Anthony W. Lin.

Career

  1. NowRPTU University Kaiserslautern-LandauPostdoctoral Researcher
  2. 2024Simons Institute for the Theory of ComputingVisiting Graduate Student

Education

  1. 2025RPTU University Kaiserslautern-LandauPhD

From the biography.

  1. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes RPTU University Kaiserslautern-Landau. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.
  2. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes University of California, Berkeley. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. The median h-index is 48 at universities and 14 at frontier AI labsIncludes University of California, Berkeley. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  4. Companies supply the most prior-work hires at 7 of the 9 frontier AI labsIncludes University of California, Berkeley. Our estimates of the organizations where 550 people worked directly before joining 9 frontier AI labs across 620 career transitions, comparing the main feeder organizations per lab.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Pascal Bergsträßer yet. Send us a free-licensed photo.

  1. [2205.09015] Ramsey Quantifiers over Automatic Structures: Complexity and Applications to Verification arxiv.org. Checked 22 October 2026.
  2. [2311.04031] Ramsey Quantifiers in Linear Arithmetics arxiv.org. Checked 22 October 2026.
  3. Pascal Bergsträßer | Simons Institute for the Theory of Computing simons.berkeley.edu. Checked 22 October 2026.
  4. Homepage of Anthony Widjaja Lin [previously Anthony W. To] anthonywlin.github.io. Checked 22 October 2026.
  5. [2510.19315] Transformers are Inherently Succinct arxiv.org. Checked 22 October 2026.
  6. [2511.05323] Fast Ramsey Quantifier Elimination in LIRA (with applications to liveness checking) arxiv.org. Checked 22 October 2026.
  7. Announcing the ICLR 2026 Outstanding Papers blog.iclr.cc. Checked 22 October 2026.
  8. Pascal Bergsträßer - RPTU arg.cs.rptu.de. Checked 22 October 2026.