The electronic density of states (DOS) is conventionally computed from a relaxed crystal structure, which is unavailable for compounds that have been neither synthesized nor cataloged. Here we introduce DOSSIER (textbfDensity textbfof textbfStates from textbfStotextbfichiometry with textbfEncoder textbfRepresentations), a chemical language model that maps elemental composition directly to this spectrum. The encoder is pretrained by cross-modal knowledge distillation from a universal machine-learning interatomic potential; the transfer lowers the error by 11% when only 1,000 training examples are available. On the Mat2Spec benchmark, DOSSIER reaches a mean absolute error of 3.76 states eV⁻¹ against 3.64 for the best structure-aware model; on an extended Materials Project dataset, the predicted spectra yield band gaps and textitd-band descriptors with useful accuracy. Screening 11,977 binary and 13,251 five-component high-entropy alloy compositions for a textitd-projected DOS resembling that of NiPt₃ places known oxygen reduction electrocatalysts near the top of the ranking.