2026
Time-Frequency Token Advantage Clipping for Training Efficient Large Reasoning Model
AAAI 2026technical
Long Chain-of-Thought (CoT) reasoning enhances large reasoning models
9 accepted papers
Long Chain-of-Thought (CoT) reasoning enhances large reasoning models
Self-critique mechanisms significantly improve the performance of language models in complex reasoning tasks by giving them the ability to correct errors, conduct induction and deduction, and switch thinking insights. However, synthetic data methods often require human-introduced errors or sampling…
Post-training processes are essential phases in grounding pre-trained language models to real-world tasks, with learning from demonstrations or preference signals playing a crucial role in this adaptation. We present a unified theoretical framework bridging Supervised Fine-Tuning (SFT) and preferenc…
Reward models (RMs) guide the alignment of large language models (LLMs), steering them toward behaviors preferred by humans. Evaluating RMs is the key to better aligning LLMs. However, the current evaluation of RMs may not directly correspond to their alignment performance due to the limited distrib…
Despite the success of reinforcement learning from human feedback (RLHF) in aligning language models with human values, reward hacking, also termed reward overoptimization, remains a critical challenge. This issue primarily arises from reward misgeneralization, where reward models (RMs) compute rew…
Adversarial detection aims to detect adversarial samples that threaten the security of deep neural networks, which is an essential step toward building robust AI systems. Density-based estimation is widely considered as an effective technique by explicitly modeling the distribution of normal data an…
Benefiting from massive corpora and advanced hardware, large language models (LLMs) exhibit remarkable capabilities in language understanding and generation. However, their performance degrades in scenarios where multiple tasks are encountered sequentially, also known as catastrophic forgetting. In…
Point cloud completion is the task of estimating the complete point cloud from the partial observation. Most of the existing methods tend to recover global shapes of 3D objects and usually lack local details. These methods rely merely on distance metrics between point sets as loss functions, which h…
Adversarial training, which minimizes the loss of adversarially perturbed examples, has received considerable attention. However, these methods require modifying all model parameters and optimizing the model from scratch, which is parameter inefficient and unfriendly to the already deployed models.…