While decoder-only LLMs excel at a vast array of natural language tasks, it suffers from an asymmetric information flow induced by causal attention: later tokens are richer in contextual grounding than earlier ones. A simple and effective remedy is prompt repetition -- just appending a second copy of prompt before generation can redistribute grounding across positions and improve reasoning performance.
arXiv:2607. 01792v1 Announce Type: cross Abstract: While decoder-only LLMs excel at a vast array of natural language tasks, it suffers from an asymmetric information flow induced by causal attention: later tokens are richer in contextual grounding than earlier ones.
arXiv:2606. 29563v1 Announce Type: cross Abstract: Large language models (LLMs) excel at complex tasks like question answering and summarization, thanks to their ability to handle long-context inputs.
arXiv:2608. 15592v1 Announce Type: new Abstract: Efficient LLM serving is often bottlenecked by the need to pad sequences to a fixed maximum length, and this wastes compute and degrades throughput.
arXiv:2608. 12385v2 Announce Type: replace Abstract: As large language models serve ever more requests, cumulative inference cost is growing relative to the one-time cost of training.
arXiv:2607. 19358v1 Announce Type: new Abstract: Recent advances in long chain-of-thought reasoning models such as DeepSeek-R1 have led to increasingly longer inference context lengths under the test-time scaling paradigm.
arXiv:2607. 27600v1 Announce Type: new Abstract: Key-value (KV) cache management through compression and eviction strategies has emerged as an important research direction in recent years.
arXiv:2607. 08399v1 Announce Type: cross Abstract: Large language models process prompts by propagating activations through dozens of layers before generating a response.
arXiv:2511. 08577v3 Announce Type: replace-cross Abstract: Improving the reasoning abilities of Large Language Models (LLMs), especially under parameter constraints, is crucial for real-world applications.
arXiv:2510. 13554v2 Announce Type: replace-cross Abstract: The reasoning pattern of Large language models (LLMs) remains opaque, and reinforcement learning (RL) typically applies uniform credit across an entire generation, blurring the distinction between pivotal and routine steps.
arXiv:2510. 13940v4 Announce Type: replace-cross Abstract: Recent progress in large language models (LLMs) has focused on test-time scaling to improve reasoning via increased inference computation, but often at the cost of efficiency.
arXiv:2604. 20920v2 Announce Type: replace Abstract: Sparse attention can reduce the cost of long-context inference, but most variants introduce new architectural components.
Large language models (LLMs) achieve remarkable performance across a wide range of tasks, but their autoregressive decoding process incurs substantial inference costs due to inherently sequential token generation. Speculative decoding addresses this bottleneck by employing a lightweight draft model to propose multiple future tokens that are subsequently verified in parallel by a larger target model.