Newest papers

Trajectory Replay: Continuous-Time Mixture Transport for Lifelong Generative Models

Alessandro De Palma, Charles Jones, Tristan Cinquin, Fabian Theis, Frederik Wenkel · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26749 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Adaptasi model generatif data berkelanjutan yang telah dilatih sebelumnya ke data yang masuk secara berurutan tanpa melupakan tugas-tugas sebelumnya adalah tantangan utama dalam pembelajaran seumur hidup. Strategi terkemuka, replay generatif, menggunakan model generatif yang dilatih pada tugas-tugas historis untuk meningkatkan pembaruan berkelanjutan sendiri dengan sampel sintetis dari pengalaman. Namun, itu membutuhkan simulasi trajektori penuh pada setiap langkah pelatihan, yang mengumpulkan kesalahan diskretisasi dan dapat merusak distribusi sejarah yang diputar ulang. Di sini, kami mengusulkan replay trajektorial, yang melewati generasi sampel dengan melatih medan kecepatan yang diperbarui untuk mencocokkan kedua kecepatan historis beku di sepanjang trajektorialnya dan bidang kecepatan tugas yang masuk, ditimbang oleh koefisien stabilitas-plastik. Melalui

Machine translation (Meta M2M100); the English original follows.

Abstract

Adapting pre-trained continuous-time generative models to sequentially incoming data without forgetting earlier tasks is a central challenge in lifelong learning. A leading strategy, generative replay, uses a generative model trained on historical tasks to augment its own continual update with synthetic samples from experience. However, it requires full trajectory simulations at every training step, which accumulate discretization errors and can degrade the replayed historical distribution. Here, we propose trajectory replay, which bypasses sample generation by training an updated velocity field to match both a frozen historical velocity along its trajectories and the incoming task's velocity field, weighted by a stability-plasticity coefficient. Through the lens of continuous-time transport, we show the resulting optimum averages the two task velocities by their respective densities and generates the mixture of their probability paths. Combined with mean velocity distillation, trajectory replay accesses historical trajectories using only a few network evaluations per continual update. Empirically, it outperforms generative replay and distillation baselines on class-incremental image generation. In two biological case studies, trajectory replay extends generative perturbation models to new contexts and experimental batches across two data modalities while preserving treatment-effect prediction.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Trajectory, Computer science, Generative grammar, Generative model, Artificial intelligence, Forgetting

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange