Publications

David Bergström, Md Fahim Sikder, Mattias Tiger, Fredrik Heintz (2026). T5Bench: Benchmarking Unconditional Time-Series Generation Beyond Fidelity and Usefulness. Under Review.

Md Fahim Sikder, Ericka Johnson, Fredrik Heintz (2026). StreamFair: Online Fair Adaptation Under Temporal Intersectional Drift. Accepted at AITC 2026.

Md Fahim Sikder (2026). FairCert: Certifiable Error Rate Fairness for ML and LLM Decision Systems. Accepted at ECAF 2026.

Md Fahim Sikder, Daniel De Leng, Fredrik Heintz (2026). FairRep: Mitigating Intersectional Bias through Fair Representation Learning. Under Review.

Md Fahim Sikder (2025). Representative Synthetic Data for Fair Decision Making. Linköping University Electronic Press, 2025.

Md Fahim Sikder, Resmi Ramachandranpillai, Daniel De Leng, Fredrik Heintz (2025). Promoting Intersectional Fairness through Knowledge Distillation. ECAI 2025.

Md Fahim Sikder, Resmi Ramachandranpillai, Fredrik Heintz (2025). TransFusion: Generating Long, High-Fidelity Time Series using Diffusion Models with Transformers. MLWA.

Md Fahim Sikder, Daniel De Leng, Fredrik Heintz (2024). Fair4Free: Generating High-fidelity Fair Synthetic Samples using Data Free Distillation. Work in Progress.

Md Fahim Sikder, Resmi Ramachandranpillai, Daniel De Leng, Fredrik Heintz (2024). FairX: A comprehensive benchmarking tool for model analysis using fairness, utility, and eXplainability. AEQUITAS, co-located with ECAI 2024.

Resmi Ramachandranpillai, Md Fahim Sikder, David Bergström, Fredrik Heintz (2024). Bt-GAN: Generating Fair Synthetic Healthdata via Bias-transforming Generative Adversarial Networks. JAIR.