RWKV achieves Transformer-level performance with linear memory complexity for long sequences. However, its inherent sequentiality limits hardware parallelism and triggers memory-bound bottlenecks via frequent off-chip weight accesses. To address these challenges, we propose HFRWKV, an FPGA-based hardware accelerator for RWKV. Within the matrix operation module, we propose a novel Δ-PoT hardware-friendly hybrid-precision quantization strategy, which enhances performance while maintaining acceptable accuracy. For the complex operations including exponentiation and division, we introduce a method featuring reusable architectures combined with lookup tables or piecewise linear approximation, which is algorithmically refined to effectively balance precision and hardware resource consumption. We adopt a fully on-chip computing system integrating parallel matrix-vector array and efficient pipelining, employing computation reordering and chunked double buffering to eliminate data bottlenecks and maximize throughput. We implement HFRWKV on the Alveo U50 and U280 platform. Experimental results show that compared to a CPU, a throughput improvement of 63.48× and an energy efficiency improvement of 139.17×. Compared to GPUs, achieves a throughput improvement of 32.33× and an energy efficiency improvement of 171.36×.

Uplift modeling estimates individual-level treatment effects for personalized decisions in applications such as marketing and healthcare. Beyond binary treatments, many applications involve multi-valued treatments. As data are split across more groups, sparsity and imbalance pose challenges. We propose CAMU, which combines (i) a Codebook-enhanced representation network that anchors representations across mini-batches to stabilize learning and facilitate distribution alignment, and (ii) a Cross-Treatment Self-Attention (CTSA) enhanced prediction network for treatment-aware knowledge sharing. We extend CAMU to CAMUER, which transfers unbiased signals from randomized controlled trial (RCT) data via attention alignment and prediction-consistency distillation. The proposed model is deployed in Kuaishou's ad-serving system, serving tens of millions of users. Offline experiments and a large-scale online A/B test demonstrate consistent gains over baselines.

Human annotations play a crucial role in evaluating the performance of GenAI models. Two common challenges in practice, however, are missing annotations (the response variable of interest) and cluster dependence among human-AI interactions (e.g., questions asked by the same user may be highly correlated). Reliable inference must address both issues to achieve unbiased estimation and appropriately quantify uncertainty when estimating average scores from human annotations. In this paper, we analyze the doubly robust estimator, a widely used method in missing data analysis and causal inference, applied to this setting and establish novel theoretical properties under cluster dependence. We further illustrate our findings through simulations and a real-world conversation quality dataset. Our theoretical and empirical results underscore the importance of incorporating cluster dependence in missing response problems to perform valid statistical inference.

In many real-world causal inference applications, the primary outcomes (labels) are often partially missing, especially if they are expensive or difficult to collect. If the missingness depends on covariates (i.e., missingness is not completely at random), analyses based on fully observed samples alone may be biased. Incorporating surrogates, which are fully observed post-treatment variables related to the primary outcome, can improve estimation in this case. In this paper, we study the role of surrogates in estimating continuous treatment effects and propose a doubly robust method to efficiently incorporate surrogates in the analysis, which uses both labeled and unlabeled data and does not suffer from the above selection bias problem. Importantly, we establish the asymptotic normality of the proposed estimator and show possible improvements on the variance compared with methods that solely use labeled data. Extensive simulations show our methods enjoy appealing empirical performance.