Abstract We develop and demonstrate methods for simulating the scattering of particle wave packets in the interacting Thirring model on digital quantum computers, with hardware implementations on up to 80 qubits. We identify low-entanglement time slices of the scattering dynamics and exploit their efficient representation by tensor networks. Circuit compression based on matrix product state techniques yields on average a reduction by a factor of 3.2 in circuit depth compared to conventional approaches, allowing longer evolution times to be evaluated with higher fidelity on contemporary quantum processors. Utilizing zero-noise extrapolation in combination with Pauli twirling, on quantum hardware we accurately simulate the full scattering dynamics on 40 qubits, and further demonstrate the tensor networks compressed state preparation on 80 qubits. Similar content being viewed by others Introduction Scattering experiments are at the heart of unraveling the internal structure of matter and the interactions between the fundamental particles. Major experimental facilities such as LHC1 and RHIC2 continue to generate valuable experimental data to test the theoretical predictions and drive the search for new physics. Concomitantly, there are significant efforts to develop analytical and numerical methods for improving our understanding of gauge field theories, which provide the theoretical framework for particle physics. In particular, Lattice Field Theory (LFT) provides a powerful tool for exploring non-perturbative regimes from first principles. Discretizing a theory on a Euclidean space-time lattice allows for applying sophisticated Monte Carlo (MC) methods that have been extremely successful for computing properties such as mass spectra, phase diagrams and many other static properties3,4. However, the conventional MC approach to LFT is not suited to directly explore dynamical problems, as it crucially relies on the formulation in Euclidean space-time, and using Minkowski space-time would lead to a sign problem preventing efficient MC sampling. While indirect approaches exist to address scattering problems with