Response Ranking with Multi-types of Deep Interactive Representations in Retrieval-based Dialogues

Building an intelligent dialogue system with the ability to select a proper response according to a multi-turn context is challenging in three aspects: (1) the meaning of a context–response pair is built upon language units from multiple granularities (e.g., words, phrases, and sub-sentences, etc.); (2) local (e.g., a small window around a word) and long-range (e.g., words across the context and the response) dependencies may exist in dialogue data; and (3) the relationship between the context and the response candidate lies in multiple relevant semantic clues or relatively implicit semantic clues in some real cases. However, existing approaches usually encode the dialogue with mono-type representation and the interaction processes between the context and the response candidate are executed in a rather shallow manner, which may lead to an inadequate understanding of dialogue content and hinder the recognition of the semantic relevance between the context and response. To tackle these challenges, we propose a representation [ K ] -interaction [ L ] -matching framework that explores multiple types of deep interactive representations to build context-response matching models for response selection. Particularly, we construct different types of representations for utterance–response pairs and deepen them via alternate encoding and interaction. By this means, the model can handle the relation of neighboring elements, phrasal pattern, and long-range dependencies during the representation and make a more accurate prediction through multiple layers of interactions between the context–response pair. Experiment results on three public benchmarks indicate that the proposed model significantly outperforms previous conventional context-response matching models and achieve slightly better results than the BERT model for multi-turn response selection in retrieval-based dialogue systems.

Runtime Analysis of Area-Efficient Uniform RO-PUF for Uniqueness and Reliability Balancing

The main issue of ring oscillator physical unclonable functions (RO-PUF) is the existence of unstable ROs in response to environmental variations. The RO pairs with close frequency differences tend to contribute bit flips, reducing the reliability. Research on improving reliability has been carried out over the years. However, it has led to other issues, such as decreasing the uniqueness and increasing the area utilized. Therefore, this paper proposes a uniform RO-PUF, requiring a smaller area than a conventional design, aiming to balance reliability and uniqueness. We analyzed RO runtimes to increase reliability. In general, our method (uniqueness = 47.48%, reliability = 99.16%) performs better than previously proposed methods for a similar platform (Altera), and the reliability is as good as the latest methods using the same IC technology (28 nm). Moreover, the reliability is higher than that of RO-PUF with challenge and response pair (CRP) enhancements. The evaluation was performed in longer runtimes, where the pulses produced by ROs exceeded the counter capacity. This work recommends choosing ranges of the runtime of RO for better performance. For the 11-stage ROs, the range should be 1.598–4.30 ms, or 6.12–8.61 ms, or 12.24–12.91 ms. Meanwhile, for the 20-stage, the range should be 2.717–8.37 ms, or 10.97–16.74 ms, or 21.93–25.10 ms.

Handling Resistance to Change When Societal and Workplace Logics Conflict

Changes in societal logics often leave firms’ policies and practices out of step. Yet when firms introduce a change that brings in a new societal logic, employees may resist, even though they personally value the change, because the incoming logic conflicts with existing organizational logics. How can change agents handle logic-based resistance to an organizational initiative that introduces a new logic? We studied elite law firms that introduced a new role into their traditional up-or-out career path in response to associates’ anonymously expressed desire for better work–life balance, which associates resisted because expressing family concerns was illegitimate within the firms. Change agents responded to three forms of resisters’ logic-based concerns—irreconcilability, ambiguity, and contradiction—with three tailored responses—redirecting, reinforcing, and reassuring—using contextually legitimate logic elements. Over time logic elements of each concern–response pair harmonized to enable individuals to enact their logics seamlessly and organizations to update the existing logic settlement to assimilate the societal change. We demonstrate that the way available logics are accessed and activated between pluralistic change agents and resisters can enable logic settlements to be updated in response to societal change. We draw insights about how logics do or do not constrain agency.

Physical Unclonable Function (PUF)-Based e-Cash Transaction Protocol (PUF-Cash)

Electronic money (e-money or e-Cash) is the digital representation of physical banknotes augmented by added use cases of online and remote payments. This paper presents a novel, anonymous e-money transaction protocol, built based on physical unclonable functions (PUFs), titled PUF-Cash. PUF-Cash preserves user anonymity while enabling both offline and online transaction capability. The PUF’s privacy-preserving property is leveraged to create blinded tokens for transaction anonymity while its hardware-based challenge–response pair authentication scheme provides a secure solution that is impervious to typical protocol attacks. The scheme is inspired from Chaum’s Digicash work in the 1980s and subsequent improvements. Unlike Chaum’s scheme, which relies on Rivest, Shamir and Adlemans’s (RSA’s) multiplicative homomorphic property to provide anonymity, the anonymity scheme proposed in this paper leverages the random and unique statistical properties of synthesized integrated circuits. PUF-Cash is implemented and demonstrated using a set of Xilinx Zynq Field Programmable Gate Arrays (FPGAs). Experimental results suggest that the hardware footprint of the solution is small, and the transaction rate is suitable for large-scale applications. An in-depth security analysis suggests that the solution possesses excellent statistical qualities in the generated authentication and encryption keys, and it is robust against a variety of attack vectors including model-building, impersonation, and side-channel variants.

Japanese compliment discourse: The process of collaborative construction

Following on Pomerantz’s (1978) study, the speech act of compliment has been studied from various aspects. Past studies have been confined largely to examination of a single compliment and its response pair. However, actual compliment speech acts rarely end after one such exchange. Therefore, this paper observes the strategic development of compliment discourse: how compliment discourse, a series of exchanges that include compliments and their responses, develops, recesses, and moves into the next topic. In analyzing the extended speech act, rather than merely compliment response pairs, this research reveals the following developmental patterns: repetition of compliment and response, followed by frame shift to close the compliment discourse. The findings indicate that people collaboratively construct compliment discourse for the purpose of facework, i.e., avoidance of self-praise and agreement with others, and respecting evaluation of evaluation.