arXiv subjects

Yichen Jiang

Publications and source records attributed to Yichen Jiang.

24 records · Page 2Linked to original sources

arXiv2019

Avoiding Reasoning Shortcuts: Adversarial Evaluation, Training, and Model Development for Multi-Hop QA

Multi-hop question answering requires a model to connect multiple pieces of evidence scattered in a long context to answer the question. In this paper, we show that in the multi-hop HotpotQA (Yang et al., 2018) dataset, the examples often contain reasoning shortcuts through which models can directly locate the answer by word-matching the question with a sentence in the context. We demonstrate this issue by constructing adversarial documents that create contradicting answers to the shortcut but do not affect the validity of the original answer. The performance of strong baseline models drops significantly on our adversarial evaluation, indicating that they are indeed exploiting the shortcuts rather than performing multi-hop reasoning. After adversarial training, the baseline's performance improves but is still limited on the adversarial evaluation. Hence, we use a control unit that dynamically attends to the question at different reasoning hops to guide the model's multi-hop reasoning. We show that this 2-hop model trained on the regular data is more robust to the adversaries than the baseline model. After adversarial training, this 2-hop model not only achieves improvements over its counterpart trained on regular data, but also outperforms the adversarially-trained 1-hop baseline. We hope that these insights and initial improvements will motivate the development of new models that combine explicit compositional reasoning with adversarial training.

cs.CL↗

arXiv2019

Explore, Propose, and Assemble: An Interpretable Model for Multi-Hop Reading Comprehension

Multi-hop reading comprehension requires the model to explore and connect relevant information from multiple sentences/documents in order to answer the question about the context. To achieve this, we propose an interpretable 3-module system called Explore-Propose-Assemble reader (EPAr). First, the Document Explorer iteratively selects relevant documents and represents divergent reasoning chains in a tree structure so as to allow assimilating information from all chains. The Answer Proposer then proposes an answer from every root-to-leaf path in the reasoning tree. Finally, the Evidence Assembler extracts a key sentence containing the proposed answer from every path and combines them to predict the final answer. Intuitively, EPAr approximates the coarse-to-fine-grained comprehension behavior of human readers when facing multiple long documents. We jointly optimize our 3 modules by minimizing the sum of losses from each stage conditioned on the previous stage's output. On two multi-hop reading comprehension datasets WikiHop and MedHop, our EPAr model achieves significant improvements over the baseline and competitive results compared to the state-of-the-art model. We also present multiple reasoning-chain-recovery tests and ablation studies to demonstrate our system's ability to perform interpretable and accurate reasoning.

cs.CL↗

arXiv2019

Toward Scalable Fully Homomorphic Encryption Through Light Trusted Computing Assistance

It has been a long standing problem to securely outsource computation tasks to an untrusted party with integrity and confidentiality guarantees. While fully homomorphic encryption (FHE) is a promising technique that allows computations performed on the encrypted data, it suffers from a significant slow down to the computation. In this paper we propose a hybrid solution that uses the latest hardware Trusted Execution Environments (TEEs) to assist FHE by moving the bootstrapping step, which is one of the major obstacles in designing practical FHE schemes, to a secured SGX enclave. TEEFHE, the hybrid system we designed, makes it possible for homomorphic computations to be performed on smaller ciphertext and secret key, providing better performance and lower memory consumption. We make an effort to mitigate side channel leakages within SGX by making the memory access patterns totally independent from the secret information. The evaluation shows that TEEFHE effectively improves the software only FHE schemes in terms of both time and space.

cs.CR↗

arXiv2018

Closed-Book Training to Improve Summarization Encoder Memory

A good neural sequence-to-sequence summarization model should have a strong encoder that can distill and memorize the important information from long input texts so that the decoder can generate salient summaries based on the encoder's memory. In this paper, we aim to improve the memorization capabilities of the encoder of a pointer-generator model by adding an additional 'closed-book' decoder without attention and pointer mechanisms. Such a decoder forces the encoder to be more selective in the information encoded in its memory state because the decoder can't rely on the extra information provided by the attention and possibly copy modules, and hence improves the entire model. On the CNN/Daily Mail dataset, our 2-decoder model outperforms the baseline significantly in terms of ROUGE and METEOR metrics, for both cross-entropy and reinforced setups (and on human evaluation). Moreover, our model also achieves higher scores in a test-only DUC-2002 generalizability setup. We further present a memory ability test, two saliency metrics, as well as several sanity-check ablations (based on fixed-encoder, gradient-flow cut, and model capacity) to prove that the encoder of our 2-decoder model does in fact learn stronger memory representations than the baseline encoder.

cs.CL↗

arXiv2014

Privacy-Preserving Vehicular Announcements Aggregation Scheme Based on Threshold Ring Signature

As the most promising application in VANETs, the vehicular announcement allows vehicles to send announcement messages about road conditions to other vehicles far away. The security requirements of reliability and privacy in the vehicular announcement are not easily achieved simultaneously due to the notorious sybil attack. In this paper, we present a novel privacy-preserving vehicular announcements aggregation scheme. The proposed scheme provides threshold authentication and flexible anonymity using message aggregation and interactive threshold ring signature which allows nondeterministic different signers to generate signature commonly in the environment not fully trusted. Different from existing works in an attack-then-trace mode, our scheme defends against the sybil attack beforehand. To our best knowledge, it is the first privacy-preserving scheme preventing the malicious users from launching the sybil attack in advance. Through extensive evaluation, we show the availability and the efficiency of the proposed scheme.

cs.CR↗

arXiv2014

An Anonymous Communication Scheme based on Ring Signature in VANETs

Vehicular ad hoc networks allow vehicles to connect themselves as networks so that cars could communicate with each other. This paper introduces an anonymous communication scheme providing integrity protection, multi-level privacy and auditability. The scheme is based on a certificateless ring signature proposed in this paper, which is contributed to reduce the length of the signature and simplify the key management. In our scheme, vehicles can compose the anonymous group without the help of road-side infrastructure or central authority. The computation overhead is close to a normal signature scheme, so it is efficient in most application scenarios. We also present a small-scale implementation to show the availability of the prototype system.

cs.CR↗