SIGformer is a novel method for sign-aware graph-based recommendation that leverages the transformer architecture to integrate both positive and negative feedback into a signed graph. The method addresses the limitations of existing approaches by incorporating two innovative positional encodings: Sign-aware Spectral Encoding (SSE) and Sign-aware Path Encoding (SPE). SSE captures the spectral properties of the signed graph, while SPE captures the path patterns, enabling the full exploitation of the graph's collaborative information. Extensive experiments on five real-world datasets demonstrate that SIGformer outperforms state-of-the-art methods, highlighting its effectiveness in leveraging negative feedback and the proposed encodings. The code for SIGformer is available at https://github.com/StupidThree/SIGformer.SIGformer is a novel method for sign-aware graph-based recommendation that leverages the transformer architecture to integrate both positive and negative feedback into a signed graph. The method addresses the limitations of existing approaches by incorporating two innovative positional encodings: Sign-aware Spectral Encoding (SSE) and Sign-aware Path Encoding (SPE). SSE captures the spectral properties of the signed graph, while SPE captures the path patterns, enabling the full exploitation of the graph's collaborative information. Extensive experiments on five real-world datasets demonstrate that SIGformer outperforms state-of-the-art methods, highlighting its effectiveness in leveraging negative feedback and the proposed encodings. The code for SIGformer is available at https://github.com/StupidThree/SIGformer.