arXiv subjects

Ritika Borkar

Publications and source records attributed to Ritika Borkar.

7 recordsLinked to original sources

arXiv2026

Cross-Model KV Cache Transfer in LLM Families: A Closed-Form Linear Mapping for Prefill Reuse

Production deployments often swap between different-sized models in a family for cost-quality cascading, mid-conversation switching, and routing, and each swap forces the receiver to repay the prefill from scratch. We propose cross-model KV cache transfer, where the receiver reuses the source's KV cache, skipping prefill. We find that cross-model KV has substantial linear structure across matched-KV pairs, where source and target share KV head count and per-head dimension. On Qwen3 14B->32B, one source layer explains 56% of variance in the target's keys and 32% in values, rising to 79% and 65% with multiple source layers. Building on this, we design a closed-form ridge mapper that operates per head and proceeds in three steps. First, for each target layer we select the top-k most predictive source layers and concatenate their KV as input. Second, we strip RoPE from the keys before mapping, so the fit is position-free and reusable across context lengths. Third, we fit ridge regression on a small calibration set of 500 FineWeb-Edu sequences of 1,024 tokens each. Surprisingly, across six pairs in three families, this linear mapper retains 73-98% of the receiver's standalone-prefill accuracy on four pairs, while two degrade sharply. A nonlinear MLP recovers up to +37 pp HellaSwag retention on the failures. The mapper runs 2.7-25x faster than re-prefill and remains stable across multi-turn handoff, making cross-model KV cache transfer practical.

cs.LG↗

arXiv2026

Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding

RL post-training of frontier language models is increasingly bottlenecked by autoregressive rollout generation, making rollout acceleration a central systems challenge. Many existing efficiency methods improve throughput by changing the rollout or optimization regime, for example, through off-policy execution, replay, or lower-precision generation. We study speculative decoding as a lossless acceleration primitive for RL rollouts that preserves the target model's output distribution. We implement speculative decoding in NeMo-RL with a vLLM backend, supporting both synchronous and asynchronous pipelines and enabling speculation during RL rollouts. This benefit is realizable across speculation mechanisms, such as pretrained MTP heads, small external draft models or even techniques such as Eagle3, which are traditionally applied after RL phase. This yields a deployment path for state-of-the-art speculative decoding inside RL training. In a reasoning post-training workload at 8B scale under synchronous RL, speculative decoding improves rollout throughput by 1.8x. Using a high-fidelity performance simulator, we project that combining speculative decoding with asynchronous RL yields up to 2.5x end-to-end training speedup at 235B scale.

cs.LG↗

arXiv2026

LatentMoE: Toward Optimal Accuracy per FLOP and Parameter in Mixture of Experts

Mixture of Experts (MoEs) have become a central component of many state-of-the-art open-source and proprietary large language models. Despite their widespread adoption, it remains unclear how close existing MoE architectures are to optimal with respect to inference cost, as measured by accuracy per floating-point operation and per parameter. In this work, we revisit MoE design from a hardware-software co-design perspective, grounded in empirical and theoretical considerations. We characterize key performance bottlenecks across diverse deployment regimes, spanning offline high-throughput execution and online, latency-critical inference. Guided by these insights, we introduce LatentMoE, a new model architecture resulting from systematic design exploration and optimized for maximal accuracy per unit of compute. Empirical design space exploration at scales of up to 95B parameters and over a 1T-token training horizon, together with supporting theoretical analysis, shows that LatentMoE consistently outperforms standard MoE architectures in terms of accuracy per FLOP and per parameter. Given its strong performance, the LatentMoE architecture has been adopted by the flagship Nemotron-3 Super and Ultra models and scaled to substantially larger regimes, including longer token horizons and larger model sizes, as reported in Nvidia et al. (arXiv:2512.20856).

cs.LG↗

arXiv2025

Efficient MoE Serving in the Memory-Bound Regime: Balance Activated Experts, Not Tokens

Expert Parallelism (EP) permits Mixture of Experts (MoE) models to scale beyond a single GPU. To address load imbalance across GPUs in EP, existing approaches aim to balance the number of tokens each GPU processes. Surprisingly, we find that this objective degrades performance rather than improving it when processing is memory-bound - a common occurrence in MoE serving, especially in the decode phase. Our analysis reveals that balancing the number of tokens processed per GPU increases the number of activated experts, exacerbating memory pressure in the memory-bound regime. We propose Minimum Expert Token ROuting, a novel token-routing algorithm for high-performance expert-parallel MoE serving in the memory-bound regime that balances the number of activated experts per GPU rather than token counts. METRO achieves near-optimal routing quality with minimal computational overhead by jointly optimizing algorithmic efficiency and leveraging the GPU's parallel processing power. To guarantee routing quality, METRO also employs a novel allGather scheme to gather global top-k knowledge, which has minimal overhead compared to conventional allToAll. Our evaluation of METRO against EPLB on both real systems (vLLM over 8 A100 GPUs) and a proprietary simulator (8-16 B200 GPUs) shows that METRO reduces decode latency by 11 - 22%, and total token throughput by 3 - 21% for Qwen3 and DeepSeek-V3 serving, where prefill and decode phases are co-deployed. In addition, by trading latency headroom for throughput, METRO improves decode throughput by up to 4.11x over EPLB at a fixed decode SLO.

cs.DC↗

arXiv2025

Helix Parallelism: Rethinking Sharding Strategies for Interactive Multi-Million-Token LLM Decoding

As LLMs scale to multi-million-token KV histories, real-time autoregressive decoding under tight Token-to-Token Latency (TTL) constraints faces growing pressure. Two core bottlenecks dominate: accessing Feed-Forward Network (FFN) weights and reading long KV caches. While Tensor Parallelism (TP) helps mitigate the cost of FFN weight reads, it does not scale well for attention. When TP width exceeds the number of KV heads, it leads to inefficient KV duplication, limits parallelism, and constrains batch size. Simultaneously, DRAM reads for long KV histories scale linearly with batch size, further capping efficiency. We introduce Helix Parallelism, a hybrid execution strategy that applies KV parallelism during attention to shard KV caches across GPUs, then reuses the same GPUs for TP in dense LLMs or TPxExpert Parallel (EP) in MoEs during FFN computation. To preserve exact attention behavior, Helix includes a lightweight communication step. To minimize the exposed communication cost, we introduce Helix HOP-B. Helix HOP-B effectively minimizes communication overhead through batchwise overlap, preserving low TTL while improving GPU efficiency. Compared to conventional parallelism approaches, Helix reduces TTL by up to 1.5x at fixed batch sizes and supports up to 32x larger batches under the same latency budget for DeepSeek-R1, pushing forward the throughput-latency Pareto on Blackwell and making real-time inference with ultra-long-sequence practical.

cs.DC↗

arXiv2025

Beyond the Buzz: A Pragmatic Take on Inference Disaggregation

As inference scales to multi-node deployments, disaggregation - splitting inference into distinct phases - offers a promising path to improving the throughput-interactivity Pareto frontier. Despite growing enthusiasm and a surge of open-source efforts, practical deployment of disaggregated serving remains limited due to the complexity of the optimization search space and system-level coordination. In this paper, we present the first systematic study of disaggregated inference at scale, evaluating hundreds of thousands of design points across diverse workloads and hardware configurations. We find that disaggregation is most effective for prefill-heavy traffic patterns and larger models. Our results highlight the critical role of dynamic rate matching and elastic scaling in achieving Pareto-optimal performance. Our findings offer actionable insights for efficient disaggregated deployments to navigate the trade-off between system throughput and interactivity.

cs.DC↗

arXiv2025

MLPerf Power: Benchmarking the Energy Efficiency of Machine Learning Systems from Microwatts to Megawatts for Sustainable AI

Rapid adoption of machine learning (ML) technologies has led to a surge in power consumption across diverse systems, from tiny IoT devices to massive datacenter clusters. Benchmarking the energy efficiency of these systems is crucial for optimization, but presents novel challenges due to the variety of hardware platforms, workload characteristics, and system-level interactions. This paper introduces MLPerf Power, a comprehensive benchmarking methodology with capabilities to evaluate the energy efficiency of ML systems at power levels ranging from microwatts to megawatts. Developed by a consortium of industry professionals from more than 20 organizations, MLPerf Power establishes rules and best practices to ensure comparability across diverse architectures. We use representative workloads from the MLPerf benchmark suite to collect 1,841 reproducible measurements from 60 systems across the entire range of ML deployment scales. Our analysis reveals trade-offs between performance, complexity, and energy efficiency across this wide range of systems, providing actionable insights for designing optimized ML solutions from the smallest edge devices to the largest cloud infrastructures. This work emphasizes the importance of energy efficiency as a key metric in the evaluation and comparison of the ML system, laying the foundation for future research in this critical area. We discuss the implications for developing sustainable AI solutions and standardizing energy efficiency benchmarking for ML systems.

cs.AR↗