Fuzz testing is crucial for identifying software vulnerabilities,with coverage-guided grey-box fuzzers like AFL and Angora excelling in broad detection.However,as the need for targeted detection grows,directed grey-bo...Fuzz testing is crucial for identifying software vulnerabilities,with coverage-guided grey-box fuzzers like AFL and Angora excelling in broad detection.However,as the need for targeted detection grows,directed grey-box fuzzing(DGF)has become essential,focusing on specific vulnerabilities.The initial seed corpus,which consists of carefully selected input samples that the fuzzer uses as a starting point,is fundamental in determining the paths that the fuzzer explores.A well-designed seed corpus can guide the fuzzer more effectively towards critical areas of the code,improving the efficiency and success of the fuzzing process.Even with its importance,much work concentrates on refining guidance mechanisms while paying less attention to optimizing the initial seed corpus.In this paper,we introduce ISC4DGF,a novel approach to generating optimized initial seed corpus for DGF using large language models(LLMs).By leveraging LLMs’deep understanding of software and refined user inputs,ISC4DGF creates a precise seed corpus that efficiently triggers specific vulnerabilities through a multi-round validation process.Implemented on AFL and tested against state-of-the-art fuzzers such as Titan,BEACON,AFLGo,FairFuzz,and Entropic using the Magma benchmark,ISC4DGF achieves a 25.03x speedup with fewer target reaches.Moreover,ISC4DGF improves target vulnerabilities detection accuracy while narrowing the detection scope and reducing code coverage.展开更多
Seed clearing is a critical stage during precision seed metering process to ensure high seed singulation.However,there is a lack of understanding of the dynamics in the seed clearing process.In this study,a model was ...Seed clearing is a critical stage during precision seed metering process to ensure high seed singulation.However,there is a lack of understanding of the dynamics in the seed clearing process.In this study,a model was developed to predict initial seed clearing angle,in the seed clearing process using vector fields.The model was applied to an existing high-speed metering device and soybean seeds,and the model was evaluated with bench testing results.Results showed that dynamic changes in forces and constraints of seeds during the seed clearing process could be abstracted as vectors,and the changes of vector directions could be described by their phase angles.The phase angles were functions of the rotational angle of the seed meter.The phase angle of the constraint boundary linearly increases with the increase of the rotational angle.The phase angle of the force fluctuates,as the rotational angle changes.Initial seed clearing angle obtained from the phase angles varies from 8°to 59°,depending on the seeder travel speed.When comparing the values of the initial seed clearing angles predicted by the model with those from the bench tests,the root mean square error(RMSE)were from 2.73 to 3.14,and the correlation(r)between predict and observer were all higher than 0.98,indicating that the model had reasonably good accuracy.展开更多
The sizes of both seed dispersers and seeds are traits that are likely to interact to influence seed fate in many synzoochoric plant species.Here,we examined whether members of a granivorous rodent community consist&#...The sizes of both seed dispersers and seeds are traits that are likely to interact to influence seed fate in many synzoochoric plant species.Here,we examined whether members of a granivorous rodent community consisting of species of different body size vary in their effectiveness as seed dispersers,and how this relationship may be altered by seed size.We marked northern red oak(Quercus rubra)acorns with plastic tags and placed them in size-selective rodent exclosures.The exclosures allowed differential access of rodent groups based on different body size:(i)small(e.g.Peromyscus spp.);(ii)small and medium(e.g.Tamias striatus);and(iii)small,medium and large(e.g.Sciurus carolinensis)species of rodents.Acorn removal did not differ among exclosure types,but more seeds were missing when removed by small rodents,probably because of larderhoarding.The treatments did not influence the relative frequency of acorn consumption.However,small rodents cached considerably fewer and partially ate more acorns than the other 2 groups.The mean dispersal distance was the longest for cages with medium openings,intermediate for cages with large openings and the shortest for cages with small openings.Acorn mass positively affected the probability of caching and this relationship was unaffected by exclosure type.In conclusion,granivorous rodents of different body sizes strongly differed in their interactions with acorns,with small rodents acting primarily as acorn predators and medium and large species contributing significantly more to dispersal of red oaks.展开更多
基金supported by the National Key Research and Development Program of China under Grant No.2021YFB3101802.
文摘Fuzz testing is crucial for identifying software vulnerabilities,with coverage-guided grey-box fuzzers like AFL and Angora excelling in broad detection.However,as the need for targeted detection grows,directed grey-box fuzzing(DGF)has become essential,focusing on specific vulnerabilities.The initial seed corpus,which consists of carefully selected input samples that the fuzzer uses as a starting point,is fundamental in determining the paths that the fuzzer explores.A well-designed seed corpus can guide the fuzzer more effectively towards critical areas of the code,improving the efficiency and success of the fuzzing process.Even with its importance,much work concentrates on refining guidance mechanisms while paying less attention to optimizing the initial seed corpus.In this paper,we introduce ISC4DGF,a novel approach to generating optimized initial seed corpus for DGF using large language models(LLMs).By leveraging LLMs’deep understanding of software and refined user inputs,ISC4DGF creates a precise seed corpus that efficiently triggers specific vulnerabilities through a multi-round validation process.Implemented on AFL and tested against state-of-the-art fuzzers such as Titan,BEACON,AFLGo,FairFuzz,and Entropic using the Magma benchmark,ISC4DGF achieves a 25.03x speedup with fewer target reaches.Moreover,ISC4DGF improves target vulnerabilities detection accuracy while narrowing the detection scope and reducing code coverage.
基金supported by the National Natural Science Foundation of China(Grant No.32372009)the National Natural Science Foundation of China(Grant No.31971802).
文摘Seed clearing is a critical stage during precision seed metering process to ensure high seed singulation.However,there is a lack of understanding of the dynamics in the seed clearing process.In this study,a model was developed to predict initial seed clearing angle,in the seed clearing process using vector fields.The model was applied to an existing high-speed metering device and soybean seeds,and the model was evaluated with bench testing results.Results showed that dynamic changes in forces and constraints of seeds during the seed clearing process could be abstracted as vectors,and the changes of vector directions could be described by their phase angles.The phase angles were functions of the rotational angle of the seed meter.The phase angle of the constraint boundary linearly increases with the increase of the rotational angle.The phase angle of the force fluctuates,as the rotational angle changes.Initial seed clearing angle obtained from the phase angles varies from 8°to 59°,depending on the seeder travel speed.When comparing the values of the initial seed clearing angles predicted by the model with those from the bench tests,the root mean square error(RMSE)were from 2.73 to 3.14,and the correlation(r)between predict and observer were all higher than 0.98,indicating that the model had reasonably good accuracy.
基金supported by(Polish)National Science Centre grant 2012/04/M/NZ8/00674.
文摘The sizes of both seed dispersers and seeds are traits that are likely to interact to influence seed fate in many synzoochoric plant species.Here,we examined whether members of a granivorous rodent community consisting of species of different body size vary in their effectiveness as seed dispersers,and how this relationship may be altered by seed size.We marked northern red oak(Quercus rubra)acorns with plastic tags and placed them in size-selective rodent exclosures.The exclosures allowed differential access of rodent groups based on different body size:(i)small(e.g.Peromyscus spp.);(ii)small and medium(e.g.Tamias striatus);and(iii)small,medium and large(e.g.Sciurus carolinensis)species of rodents.Acorn removal did not differ among exclosure types,but more seeds were missing when removed by small rodents,probably because of larderhoarding.The treatments did not influence the relative frequency of acorn consumption.However,small rodents cached considerably fewer and partially ate more acorns than the other 2 groups.The mean dispersal distance was the longest for cages with medium openings,intermediate for cages with large openings and the shortest for cages with small openings.Acorn mass positively affected the probability of caching and this relationship was unaffected by exclosure type.In conclusion,granivorous rodents of different body sizes strongly differed in their interactions with acorns,with small rodents acting primarily as acorn predators and medium and large species contributing significantly more to dispersal of red oaks.
