The Annual Conference on Neural Information Processing Systems (NeurIPS) is one of the most prestigious and competitive international conferences in machine learning and computational neuroscience. At NeurIPS 2026, 45 of the accepted papers involves WASP PhD students, WASP alumni, supervisors and/or WASP faculty as authors or co-authors.
This number is a true success for WASP and a mark of excellence. We’re especially proud to note that this represents an increase of 18 papers compared to last year, reflecting the growing impact of WASP researchers on the global AI stage.
NeurIPS 2026 accepted 7,900 papers out of 30,709 submissions, for an acceptance rate of 25.7%.
The list of accepted papers was updated October 26, 2026.
Do you miss your paper on this list? Please email WASP.Newsletter@partner.liu.se and we will add it to the list.
WASP-related papers and posters accepted to NeurIPS 2026
Efficient LLM adaptation with forward-only passes
B. Huang, A. Balashankar, A. Aminifar
ESS-Flow: Training-free guidance of flow-based models as inference in source space
A. Kalaivanan, Z. Zhao, J. Sjölund, F. Lindsten
Njord: A probabilistic graph neural network for ensemble ocean forecasting
D. Holmberg, J. Oskarsson, E. Wikingsson, F. Lindsten, T. Roos
Generating symmetric crystals using latent flow matching
A. Karmush, C. M. Brandenburg, S. Ershadrad, J. Rosén, M. Felsberg, F. Ekström Kelvinius
The multi-block DC function class: Theory, algorithms, and applications
P. Fatemi, H. Maskan, A. Yurtsever, S. Sra
Tight generalization bounds for noiseless inverse optimization
P. Fatemi, H. Maskan, S. Sra, P. Mohajerin Esfahani
Anytime-valid conformal risk control
B. Hultberg, D. Zachariah, A. H. Ribeiro
Learning inexact alternating minimization for convex problems with quadratic coupling
P. Häusner, J. Rudzusika, J. Sjölund, O. Öktem
Self-rewarding sequential Monte Carlo for masked diffusion language models
Z. Luo, Z. Jin, L. Wang, L. Bing, T. B. Schön
TIDE: Asymmetric neural circuits for stabilized temporal inhibitory-excitatory dynamics
A. Kyuroson, D. Kleyko, M. Liwicki
DiA: Directional adapter
P. C. Chhipa, A. Singh
QT-Net: Rethinking evaluation of AI models in atomic chemical space
P. Martínez Crespo, S. Ribes, M. Rahm, R. J. M. Beckmann, R. Jordan, M. Gliege, S. Miret, V. K. Narasimhan, R. Mercado
Adaptive prior selection in Gaussian process bandits with Thompson sampling
J. Sandberg, M. Haghir Chehreghani
The interplay of data structure and imbalance in the learning dynamics of diffusion models
F. Nicoletti, C. Ma, E. Ventura, L. Saglietti, S. Sarao Mannelli
Reflected Schrödinger bridge matching
M. Häggbom, V. Nilsson, P. Nyquist, J. Andén
When graph structure provably helps classification: Non-asymptotic recovery guarantees
F. Shahriari-Mehr, J. Aliakbari, A. Graell i Amat, A. Panahi
Deep probabilistic supervision for image classification
A. Adelöw, M. Gamba, A. Maki
State of thought enables endogenous reasoning
Z. Gong, Y. Hou, Z. Zeng, M. Xiao, C. Yuen, W. Y. B. Lim
Extracting training data from diffusion language models via infilling
Y. Wang, N. Asokan
The adversarial gait: Detecting visual adversarial attacks against vision-language models via self-targeted gradient characterization
M. Byrd Victorica, E. Shereen, G. Dán
Instance-optimal estimation with multiple LLM judges on a budget
J. Lee, S. Kim, Y. Jedra, A. Proutiere, S. Yun
Parabolic position encoding: Vision-centric, principled, extrapolatable, general
C. K. Ohrstrom, R. I. C. Muchacho, Y. Dong, F. Moumtzidellis, R. Güldenring, F. T. Pokorny, L. Nalpantidis
Bridging structure and language: Graph-based visual reasoning for autonomous road understanding
L. Wild, K. Luo, M. Pavone
Exact power indices for plurality-voting ensembles
I. Sarpe, T. Georgakopoulos, A. Gionis
A matter of interest: Understanding interestingness judgments of math problems in humans and language models
S. Mishra, Y. Machino, G. Poesia, A. Q. Jiang, J. Hsu, A. Weller, C. Mishra, D. Broman, J. Tenenbaum, M. Jamnik, C. Zhang, K. Collins
Density-ratio losses for post-hoc learning to defer
A. Soen, R. Thobaben, J. Jaldén, R. Nock
LITE: A lightweight lazy sampler for efficient SGD
A. Daghestani, M. Johansson
Sparsely supervised diffusion
W. Zhao, Z. Li, Y. Zhao, M. Vali, M. Trapp, J. Pajarinen, J. Kannala, A. Solin
Differentiable systematic resampling for variational sequential Monte Carlo
F. Cumlin, S. Chatterjee
Integrating local and global entropy for uncertainty quantification in LLMs
J. Medina, T. Zhou, K. Isufaj, A. Gionis, S. Chawla
Speeding up log-sum-exp: Kernel fusion at the memory wall, integer arithmetic at the compute wall
L. Yao, M. Andraud, N. Loppi, A. Pilzer, A. Liu, G. Van den Broeck, M. Trapp
Robust flow matching under target corruption and label noise
M. C. Kurucu, E. Englesson, H. Azizpour
Calibrated target noise recovers curvature from the gradients
A. Jamshidi, K. Haitsiukevich, A. Gionis, K. Puolamäki
Polar transformers: -Equivariant angular attention for Cryo-EM image-set processing
J. Andén, J. Sagemüller
Beyond Chamfer distance: Granular order-aware evaluation metric for online mapping
C. Bencheikh Lehocine, A. Lilja, J. Fu, L. Hammarstrand
Looking under the streetlight: Evaluation in generative molecular dynamics
S. Olsson, F. Noe, G. Rotskoff, K. Lindorff-Larsen
Efficient lookahead encoding and abstracted width for learning general policies in classical planning
M. Aichmüller, S. Ståhlberg, M. Funkquist, H. Geffner
Learning to deaggregate: Large-scale trajectory generation with spatial priors
D. Bergström, M. Tiger, F. Heintz
Geometric instability of hidden-state trajectories predicts reasoning failures in large language models
H. Fakharzadehjahromy, A. C. Bueff, F. Heintz, M. Tiger, Q. Li, J. Li, J. Geng
Reinforcement learning for long-horizon unordered tasks: From Boolean to coupled reward machines
K. Levina, N. Pappas, A. Karapantelakis, A. Vulgarakis Feljan, J. Seipp
Property-guided LLM program synthesis for planning
A. G. Pereira, A. B. Corrêa, J. Seipp
Constrained factorization with diagonal scaling: Rank-revealing training and pruning
Y. Hou, E. Akbas, S. Sra, A. Yurtsever
AcceleGrad#: Adaptive geometry-aware acceleration
H. Goralija, F. Tonin, K. Antonakopoulos, A. Yurtsever, V. Cevher
RAIL: Representation-aligned imitation learning for student-compatible teacher policies
M. Mammadov, P. Zuidberg Dos Martires, J. A. Stork
HeatKV: Head-tuned KV-cache compression for visual autoregressive modeling
J. Cederlund, A. Berg, D. A. E. Acar, C. Zhou, P. Giselsson
Published: October 26th, 2026
[addtoany]