The self-assembled quantum dots (QDs) provide an ideal platform for realization of quantum information technology because it provides on demand single photons, entangled photon pairs from biexciton cascade process, single spin qubits, and so on. The fine structure splitting (FSS) of exciton is a fundamental property of QDs for thees applications. From the symmetry point of view, since the two bright exciton states belong to two different representations for QDs with C2v symmetry, they should not only have different energies, but also have different lifetimes, which is termed exciton lifetime asymmetry. In contrast to extensively studied FSS, the investigation of the exciton lifetime asymmetry is still missed in literature. In this work, we carried out the first investigation of the exciton lifetime asymmetry in self-assembled QDs and presented a theory to deduce lifetime asymmetry indirectly from measurable qualities of QDs. We further revealed that intrinsic lifetimes and their asymmetry are fundamental quantities of QDs, which determine the bound of the extrinsic lifetime asymmetries, polarization angles, FSSs, and their evolution under uniaxial external forces. Our findings provide an important basis to deeply understanding properties of QDs.PACS numbers: 78.67. Hc, 73.21.La The self-assembled quantum dots (QDs) provide a promising platform for realizing on-demand entangled photon pairs from the biexciton-exciton-vacuum cascade process [1], which are essential for practical quantum communication [2][3][4][5][6][7]. However the major obstacle in realizing this goal comes from the non-degeneracy of the two intermediate bright exciton states (see Fig. 1a), in which their energy difference, called fine structure splitting (FSS), is much larger than the homogeneous broadening of the emission lines (Γ ∼ 1 µeV[8-10]), thus the "which-way" information is erased and only classically corrected photons instead of maximally entangled photon pairs can be created from this cascade process. In the past decade, strenuous efforts have been devoted to eliminate this splitting by applying various experimental techniques, including thermal annealing [11][12][13][14][15], electric field[16-23], magnetic field [24][25][26][27] and external stress [28][29][30][31][32][33][34][35][36][37][38] etc.. However, none of them is efficient. In recent years, the entangled photon pairs were demonstrated in a way that first picking out the QDs with small FSS from a QDs ensemble after post annealing and then eliminating the FSS using magnetic fields (see more details in the first experiment by Stevenson et al. in 2006[27]). It is worth to note that only a tiny fraction of QDs in experimental grown samples can be used to achieve entangled photon pairs. Moreover, these devices can only work at low temperature since the emitted exciton energies are very close to the emission lines from the wetting layer [39]. The mechanism underlying the difficulty of eliminating the FSS comes fun-damentally from the low symmetry of self-assembled QDs. The high...