Continuous depth batching (CDB) is introduced, which schedules at the granularity of individual loop iterations, and handles boundary stages and loop steps in separate priority queues, makes exit decisions one step ahead, and overlaps all scheduling work with GPU computation.
Abstract
A main promise of looped language models (LMs) is depth-adaptive inference. By iterating a block of shared layers a variable number of times, the model can use less compute for"easy"tokens and more for"hard"ones. However, this adaptivity breaks standard batching: tokens in the same batch now require a different number of loops, so there is no unified forward pass, making efficient inference difficult. Standard inference frameworks like vLLM schedule on the token level and cannot handle this because tokens need to be removed from the batch within the forward pass. Loop-level scheduling has been proposed as a solution, but never implemented end to end. The key challenge is that looped architectures also contain non-looped boundary stages (e.g., token embedding and LM head) that must be scheduled at different frequencies than the loop. We introduce continuous depth batching (CDB), which schedules at the granularity of individual loop iterations. CDB handles boundary stages and loop steps in separate priority queues, makes exit decisions one step ahead, and overlaps all scheduling work with GPU computation. On Ouro 1.4B and Huginn 3.5B, CDB can realize up to $99\%$ of the theoretical maximum speed-up from adaptive-depth, translating to $1.5$-$1.9\times$ higher offline throughput and $45$-$90\%$ lower normalized latency under dynamic serving load.
Recent work on looped language models suggests that many reasoning problems benefit from greater computational depth rather than from additional independent parameters. Existing studies, however, focus almost exclusively on Transformer backbones, leaving open whether this principle also applies to state-space language models. We investigate Looped Mamba and Looped Hybrid Mamba-Transformer architectures, which repeatedly apply a shared Mamba (or hybrid) block to introduce explicit finite-depth recurrent computation. On two controlled reasoning tasks-Mano (modular-arithmetic manipulation) and p-hop induction-Looped Mamba consistently outperforms parameter-matched non-looped baselines and, in several settings, matches or exceeds non-looped models of equal effective depth. We then extend the study to language model pre-training under matched iso-parameter and iso-FLOPs protocols, which jointly disentangle the effects of parameter sharing and effective depth: looped models remain competitive on downstream benchmarks with substantially fewer distinct parameters, although deeper non-looped models retain an advantage in validation perplexity under strict iso-FLOPs comparisons. Finally, we adapt Ouro's two-stage exit gate to Looped Mamba for threshold-controlled selection among recurrent-step outputs. Executing such exits on a state-space backbone, however, leaves the recurrent state without its deeper updates, and validation perplexity then degrades severely. We therefore introduce a cache-hole adaptation that aligns continued training with skipped-state inference. At the scales studied, the adapted model keeps perplexity close to full computation and matches or exceeds full-compute exit-state selection on downstream benchmarks while executing roughly half of the recurrent steps, which translates into measured inference speedups once the prefill is compute-bound.
Zhen-Xuan Yu, Takeshi Kojima, Yutaka Matsuo et al.· 0 citations
Repeating a small block of middle layers increases a language model's effective inference depth without adding parameters or generating extra tokens, and recent work shows that this latent recurrence improves reasoning. However, two design choices limit these gains. Each iteration sees only the previous output and cannot directly access earlier computations. Moreover, a fixed loop count wastes depth on easy inputs while leaving hard ones with too little computation. We introduce RecurTrace, which addresses both limitations using the loop's own trajectory. Specifically, Loop Memory Attention lets each looped layer attend to its own states from previous iterations along the loop-time axis, so the model can revisit earlier computations instead of relying on the latest state alone. A halting head then reads the loop state and predicts whether to continue, with supervision from an oracle that identifies when additional depth still reduces loss. In a controlled MathQA comparison on the same looped backbone, RecurTrace achieves 56.9% accuracy with an average of 2.0 loops, exceeding the best fixed loop depth by 2.2 points at matched compute. By comparison, ACT and PonderNet collapse to one loop, and CALM reaches only 54.1% with 5.6 loops, while the stronger LoopUS-Conf and TaH-Mismatch baselines reach 55.3% at 3.2 loops and 55.7% at 2.1 loops. Finally, RecurTrace improves generation accuracy over same-budget fine-tuned baselines at 0.6B, 1.7B, 4B, and 8B, with the gain growing with model size from 0.6 to 3.4 points.
Yu-Xiang Wang, Kun-Yu Feng, Ying-Da Shen et al.· 0 citations
CoRun is presented, a scheduling-based system that achieves deterministic inference without requiring batch invariance, and employs isolated prefill and fixed-shape batched decode to handle the two stages of LLM inference, respectively, leveraging CUDA graphs for efficient execution and simplified implementation.
Shiju Zhao, Jiacheng Yang, Qihang Chen et al.· 0 citations
Speculative reduction is proposed, which initiates data transfer before the top barrier and ensures correctness via lightweight validation during low-latency inference, which reduces synchronization overhead during low-latency inference.
Hritvik Taneja, A. Saxena, Abhishek Revinipati et al.· arXiv.org· 1 citation
Large Language Models (LLMs) trained using Chain-of-Thought (CoT) supervision have achieved state-of-the-art performance on complex reasoning tasks. However, the generation of long reasoning chains introduces substantial computational overhead during inference, limiting their deployment in low-latency and resource-constrained environments. This paper proposes AdaptiReason, a novel framework that dynamically compresses intermediate reasoning steps based on task difficulty and model confidence without requiring retraining of the underlying base model. AdaptiReason employs a lightweight difficulty estimator to determine the appropriate reasoning depth for each input, followed by a learned token-pruning policy that eliminates redundant or low-information reasoning steps. Experimental evaluation on the MATH, GSM8K, and ARC-Challenge benchmarks demonstrates that AdaptiReason reduces the average number of generated tokens by 3.7× while preserving 98.2% of the baseline reasoning accuracy. Furthermore, the proposed framework is model-agnostic and can be seamlessly integrated with instruction-tuned LLMs without requiring access to model parameters, relying solely on output logits for adaptive reasoning compression. The results demonstrate that AdaptiReason significantly improves inference efficiency while maintaining high reasoning performance, making it suitable for real-time and resource-constrained LLM applications.
V. A, Mithaguru, Amrita Kundu et al.· 2026 4th International Confe...· 0 citations
This work proposes Chained Recursive Language Models (Chained RLM), an inference-time architecture, in which the same underlying model is called repeatedly as a sequence of fresh reasoning roots, and studies when fresh-context artifact continuation gives a measurable gain in accuracy over direct LLM answering even with recursive tool-calling.
Purbesh Mitra, S. Ulukus· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
