I do not believe there are any other methods you could use that wouldn't ultimately harm the Tor network in some way.
Limiting the Bandwidth, and setting an Exit Policy are the only tools that Tor provides to configure your exit. Any other policy would have to either
Use some form of traffic inspection, to determine if the traffic is "malicious", for whatever definition of malicious you use. This is liable to get you marked as a bad exit, as things like dropping specific connections are detectable.
Apply to all traffic indiscriminately, which thus will also affect the "legitimate" traffic that you don't wish to affect.
Finally, I don't think the conception of "malicious" that most people have with reference to Tor is easy to detect anyway. As an example, one of the often cited malicious uses are people sending traffic designed to penetrate secure services, or firewalls. Yet that traffic could be coming from someone inside the network, whose job is to perform "penetration testing" on the network.
Another example is people using Tor to obtain illegal material. Yet the person obtaining the material may be a police officer building a case and collecting evidence, who doesn't want an IP associated with their organisation showing up as a red flag to the person sharing the material.