Speculative sampling is a promising approach to accelerate the decoding stage for Large Language Models (LLMs). Recent advancements that leverage target LLM's contextual information, such as hidden states and KV cache, have shown significant practical improvements. However, these approaches suffer from inconsistent context between training and decoding. We also observe another discrepancy between the training and decoding objectives in existing speculative sampling methods. In this work, we propose a solution named HArmonized Speculative Sampling (HASS) that learns harmonized representations to address these issues. HASS accelerates the decoding stage without adding inference overhead through harmonized objective distillation and harmonized context alignment. Experiments on four LLaMA models demonstrate that HASS achieves 2.81x-4.05x wall-clock time speedup ratio averaging across three datasets, surpassing EAGLE-2 by 8%-20%. The code is available at https://github.com/HArmonizedSS/HASS.

Trigger-action programming (if-this-then-that rules) empowers non-technical users to automate services and smart devices. As a user’s set of trigger-action programs evolves, the user must reason about behavior differences between similar programs, such as between an original program and several modification candidates, to select programs that meet their goals. To facilitate this process, we co-designed user interfaces and underlying algorithms to highlight differences between trigger-action programs. Our novel approaches leverage formal methods to efficiently identify and visualize differences in program outcomes or abstract properties. We also implemented a traditional interface that shows only syntax differences in the rules themselves. In a between-subjects online experiment with 107 participants, the novel interfaces better enabled participants to select trigger-action programs matching intended goals in complex, yet realistic, situations that proved very difficult when using traditional interfaces showing syntax differences.

End-user programming, particularly trigger-action programming (TAP), is a popular method of letting users express their intent for how smart devices and cloud services interact. Unfortunately, sometimes it can be challenging for users to correctly express their desires through TAP. This paper presents AutoTap, a system that lets novice users easily specify desired properties for devices and services. AutoTap translates these properties to linear temporal logic (LTL) and both automatically synthesizes property-satisfying TAP rules from scratch and repairs existing TAP rules. We designed AutoTap based on a user study about properties users wish to express. Through a second user study, we show that novice users made significantly fewer mistakes when expressing desired behaviors using AutoTap than using TAP rules. Our experiments show that AutoTap is a simple and effective option for expressive end-user programming.

Energy harvesting is a promising solution to power ubiquitous Internet-of-Things (IoT) devices. But the frequent and inevitable power failure incurs significant backup overhead, greatly degrading performance and energy efficiency. Nonvolatile processor (NVP), which can checkpoint processor states, is designed to tackle this problem. The conventional system-level design method involves repeated system modification and verification on hardware, in which measurement on hardware consumes the majority time. To expedite the NVP-based system design process, we propose a rapid system prototyping flow to eliminate repeated hardware measurement in the design flow. This method involves an NVP system-level simulator, which takes the harvester power trace, system characteristics extracted from hardware, and user design as the input, and analyzes system energy and time profile under this power trace. Iterative system optimization and verification are conducted on the simulator, with only the final verification on hardware. We demonstrate the advantages of this method by two design cases, in which time, energy efficiency and the impact of different capacitor size are optimized.