Tracked topic
…This training-time attention dilution (the starvation of content tokens in the attention distribution) weakens the gradient signal, limiting the model's ability to learn robust long-context capabilities. We introduce FocuSFT…
…Existing safety-alignment signals are largely response-level or off-policy, and often incur a safety-utility trade-off: improving agent safety comes at the cost of degraded task performance . Such sparse…
…The prediction error provides a novelty signal: familiar semantic trajectories become easier to predict, while under-explored directions produce higher error. We then use this signal to guide sampling toward less redundant…
…We focus on predictive tasks where the modalities provide complementary predictive signal , and where generic embeddings lose critical information, necessitating Target-Aware Representations that are aligned with the task. Our experimental results…
…some require costly retraining to accommodate variable-length outputs, while others depend solely on local confidence signals during decoding. Such local criteria fail to capture the evolving structure of the sequence, often…
…Generate proposes candidate trajectories and topologies; Filter constructs intermediate signals via verifiers , judges , critics ; Control allocates compute and makes continuation/branching/stopping decisions under budgets; and Replay retains and reuses artifacts across…
…LEAD dynamically calibrates the correctness-efficiency trade-off at each step using a Potential-Scaled Instability , directing optimization capacity to the most informative learning signal. Furthermore, it estimates an adaptive per-problem…
…regularization, an auxiliary policy gradient loss anchored to clean-image advantages preserves a reliable reward signal; and to avoid systematically incorrect invariance, correctness-conditioned regularization restricts enforcement to successful trajectories. On Qwen3…
…Our theory-grounded analysis shows that patch-wise modeling yields provably larger discrepancies when localized forensic signals are present in generated images, enabling more reliable separation from real images. Extensive experiments demonstrate…
…This yields a richer and more diverse learning signal , improving exploration efficiency from imperfect model behavior. As a result, DenoiseRL improves reasoning performance and overall training efficiency while reducing the need for…