Code obfuscation is a crucial technique for protecting software against reverse engineering and security attacks.Among various obfuscation methods,opaque predicates,which are recognized as flexible and promising,are w...Code obfuscation is a crucial technique for protecting software against reverse engineering and security attacks.Among various obfuscation methods,opaque predicates,which are recognized as flexible and promising,are widely used to increase control-flow complexity.However,traditional opaque predicates are increasingly vulnerable to Dynamic Symbolic Execution(DSE)attacks,which can efficiently identify and eliminate them.To address this issue,this paper proposes a novel approach for anti-DSE opaque predicates that effectively resists symbolic execution-based deobfuscation.Our method introduces two key techniques:single-way function opaque predicates,which leverage hash functions and logarithmic transformations to prevent constraint solvers from generating feasible inputs,and path-explosion opaque predicates,which generate an excessive number of execution paths,overwhelming symbolic execution engines.To evaluate the effectiveness of our approach,we implemented a prototype obfuscation tool and tested it against prominent symbolic execution engines.Experimental results demonstrate that our approach signifi-cantly increases resilience against symbolic execution attacks while maintaining acceptable performance overhead.This paper provides a robust and scalable obfuscation technique,contributing to the enhancement of software protection strategies in adversarial environments.展开更多
Modern compression and acceleration methods for exploring efficient deep neural networks render real-world applications more feasible.Existing approaches uniformly apply the same procedure to every input image,overloo...Modern compression and acceleration methods for exploring efficient deep neural networks render real-world applications more feasible.Existing approaches uniformly apply the same procedure to every input image,overlooking instancewise complexity variations.Moreover,owing to pruning or decomposition techniques,the upper bound of network representation capabilities might be permanently diminished.In this work,an input-dependent multiscale dynamic inference method(MSDI)is developed to strike a better balance between model performance and inference acceleration.Specifically,we modify the main body of a convolutional network to obtain a series of parameter-sharing subnetworks with varying levels of complexity.A side branch structure is then introduced to assign an input instance to a suitable subnetwork as its inference route,and we expect to accelerate the inference by assigning the easy input to the subnetwork with low capacity.We further propose multiscale distillation training to optimize the training of the modified subnetworks.Additionally,we compare the entropy-based and learning-based grading approaches,aiming to obtain a more suitable route assignment method.Experiments show that MSDI can accelerate most existing convolutional models,achieving up to 74.7%computation savings across diverse datasets.展开更多
The paper presents an dynamic execution model of complex real-time software based on requirement description model RTRSM, and then propose a checking method based on configuration covering and its corresponding algori...The paper presents an dynamic execution model of complex real-time software based on requirement description model RTRSM, and then propose a checking method based on configuration covering and its corresponding algorithm. This checking method can check the execution situations between parallel elements in a dynamic execution step of real-time software systems. It also can check all the states and transitions which assure the completeness of checking. In the end, related theorem is proofed.展开更多
Recent advancements in large language models(LLMs)have significantly contributed to the progress of the Text-to-SQL task.A common requirement in many of these works is the post-correction of SQL queries.However,the ma...Recent advancements in large language models(LLMs)have significantly contributed to the progress of the Text-to-SQL task.A common requirement in many of these works is the post-correction of SQL queries.However,the majority of this process entails analyzing error cases to develop prompts with rules that eliminate model bias.And there is a weakness of execution verification for SQL queries.In addition,the prevalent techniques primarily depend on GPT-4 and few-shot prompts,resulting in expensive costs.To investigate the effective methods for SQL refinement in a cost-efficient manner,we introduce Semantic-Enhanced Text-to-SQL with Adaptive Refinement(SEA-SQL),which includes Adaptive Bias Elimination and Dynamic Execution Adjustment,aims to improve performance while minimizing resource expenditure with zero-shot prompts.Specifically,SEA-SQL employs a semantic-enhanced schema to augment database information and optimize SQL queries.During the SQL query generation,a fine-tuned adaptive bias eliminator is applied to mitigate inherent biases caused by the LLM.The dynamic execution adjustment is utilized to guarantee the executability of the bias eliminated SQL query.We conduct experiments on the Spider and BIRD datasets to demonstrate the effectiveness of this framework.The results demonstrate that SEA-SQL achieves state-of-the-art performance in the GPT-3.5 scenario with 9%-58% of the generation cost.Furthermore,SEA-SQL is comparable to GPT-4 with only 0.9%-5.3% of the generation cost.Our code is available at the website of github.com/545999961/SEA-SQL.展开更多
基金supported byOpen Foundation of Key Laboratory of Cyberspace Security,Ministry of Education of China(No.KLCS20240211)Henan Science and Technology Major Project No.241110210100.
文摘Code obfuscation is a crucial technique for protecting software against reverse engineering and security attacks.Among various obfuscation methods,opaque predicates,which are recognized as flexible and promising,are widely used to increase control-flow complexity.However,traditional opaque predicates are increasingly vulnerable to Dynamic Symbolic Execution(DSE)attacks,which can efficiently identify and eliminate them.To address this issue,this paper proposes a novel approach for anti-DSE opaque predicates that effectively resists symbolic execution-based deobfuscation.Our method introduces two key techniques:single-way function opaque predicates,which leverage hash functions and logarithmic transformations to prevent constraint solvers from generating feasible inputs,and path-explosion opaque predicates,which generate an excessive number of execution paths,overwhelming symbolic execution engines.To evaluate the effectiveness of our approach,we implemented a prototype obfuscation tool and tested it against prominent symbolic execution engines.Experimental results demonstrate that our approach signifi-cantly increases resilience against symbolic execution attacks while maintaining acceptable performance overhead.This paper provides a robust and scalable obfuscation technique,contributing to the enhancement of software protection strategies in adversarial environments.
文摘Modern compression and acceleration methods for exploring efficient deep neural networks render real-world applications more feasible.Existing approaches uniformly apply the same procedure to every input image,overlooking instancewise complexity variations.Moreover,owing to pruning or decomposition techniques,the upper bound of network representation capabilities might be permanently diminished.In this work,an input-dependent multiscale dynamic inference method(MSDI)is developed to strike a better balance between model performance and inference acceleration.Specifically,we modify the main body of a convolutional network to obtain a series of parameter-sharing subnetworks with varying levels of complexity.A side branch structure is then introduced to assign an input instance to a suitable subnetwork as its inference route,and we expect to accelerate the inference by assigning the easy input to the subnetwork with low capacity.We further propose multiscale distillation training to optimize the training of the modified subnetworks.Additionally,we compare the entropy-based and learning-based grading approaches,aiming to obtain a more suitable route assignment method.Experiments show that MSDI can accelerate most existing convolutional models,achieving up to 74.7%computation savings across diverse datasets.
基金Supported by the Natural Science Foundation ofHubei Province (99J165)
文摘The paper presents an dynamic execution model of complex real-time software based on requirement description model RTRSM, and then propose a checking method based on configuration covering and its corresponding algorithm. This checking method can check the execution situations between parallel elements in a dynamic execution step of real-time software systems. It also can check all the states and transitions which assure the completeness of checking. In the end, related theorem is proofed.
基金supported by the National Natural Science Foundation of China(Grant Nos.62272054,62192784,62172056)the Beijing Nova Program(No.20230484319)the Xiaomi Young Talents Program.
文摘Recent advancements in large language models(LLMs)have significantly contributed to the progress of the Text-to-SQL task.A common requirement in many of these works is the post-correction of SQL queries.However,the majority of this process entails analyzing error cases to develop prompts with rules that eliminate model bias.And there is a weakness of execution verification for SQL queries.In addition,the prevalent techniques primarily depend on GPT-4 and few-shot prompts,resulting in expensive costs.To investigate the effective methods for SQL refinement in a cost-efficient manner,we introduce Semantic-Enhanced Text-to-SQL with Adaptive Refinement(SEA-SQL),which includes Adaptive Bias Elimination and Dynamic Execution Adjustment,aims to improve performance while minimizing resource expenditure with zero-shot prompts.Specifically,SEA-SQL employs a semantic-enhanced schema to augment database information and optimize SQL queries.During the SQL query generation,a fine-tuned adaptive bias eliminator is applied to mitigate inherent biases caused by the LLM.The dynamic execution adjustment is utilized to guarantee the executability of the bias eliminated SQL query.We conduct experiments on the Spider and BIRD datasets to demonstrate the effectiveness of this framework.The results demonstrate that SEA-SQL achieves state-of-the-art performance in the GPT-3.5 scenario with 9%-58% of the generation cost.Furthermore,SEA-SQL is comparable to GPT-4 with only 0.9%-5.3% of the generation cost.Our code is available at the website of github.com/545999961/SEA-SQL.
