Long Chain-of-Thought (CoT) reasoning improves LLM problem-solving but is computationally expensive due to sequential token generation. While recent works explore reasoning in continuous latent spaces to bypass discrete token generation, they often struggle with training stability and fail to scale to complex, long-horizon tasks due to lack of supervision signal. We propose SuperThoughts, which compresses pairs of consecutive CoT tokens into single latent representations and decodes two tokens per step via a lightweight Multi-Token Prediction (MTP) module. This preserves discrete token supervision at training time while doubling throughput at inference time. We finetune Qwen2.5-Math-1.5B-Instruct, Qwen2.5-Math-7B-Instruct, Qwen2.5-Math-14B-Instruct, and evaluate on MATH500, AMC, OlympiadBench, and GPQA-Diamond. With a confidence-based adaptive mechanism that falls back to standard decoding when uncertain, SuperThoughts achieves $\sim$20--30\% CoT length reduction while maintaining accuracy with minimal degradation (1-2 points accuracy drop on most tasks).

翻译：长链式思维推理虽能提升大语言模型的问题求解能力，但顺序令牌生成机制导致计算成本高昂。尽管近期研究探索在连续潜空间中进行推理以规避离散令牌生成，但此类方法常面临训练稳定性不足的问题，且因缺乏监督信号而难以扩展到复杂长程任务。我们提出SuperThoughts，该方法将连续成对的CoT令牌压缩为单一潜表示，并通过轻量级多令牌预测模块实现每步解码双令牌。该方案在训练阶段保持离散令牌监督信号，同时在推理阶段将吞吐量提升一倍。我们在Qwen2.5-Math-1.5B-Instruct、Qwen2.5-Math-7B-Instruct、Qwen2.5-Math-14B-Instruct模型上微调，并在MATH500、AMC、OlympiadBench和GPQA-Diamond基准上评估。结合基于置信度的自适应机制（在不确定时退化为标准解码），SuperThoughts实现了约20-30%的CoT长度缩减，同时保持准确性且性能退化极小（多数任务准确率下降仅1-2个百分点）。