Ideally you’re getting product-specific EPDs, so it’d be specific to that plant and supply chain, though industry-average EPDs or material baseline values are commonly used in LCA. I haven’t looked at many steel EPDs, but I suspect if it’s coming from an EAF they would want to brag about that in their product description in the EPD.
We’re lucky in the US that EAF’s are the dominant producers of steel (about 2/3). Regardless, an easy proxy for EAF vs BOF is specifying certain shapes that typically come from EAFs (hot rolled shapes like W’s, L’s, channels, rebar) vs shapes that come from BOFs (HSS, deck).
For steel from a manufacturing perspective, there are a lot of strategies in the works - hydrogen or biofuel replacement of fossil fuels, carbon capture and storage retrofits of steel production, more electrification (EAFs) and decarbonizing the grid, early retirement of BOFs and biomass substitution for coke for existing BOFs/DRI. But these are often costly or difficult, so without strong policy incentive and pending further R&D, it may be tough to shift the practice. Using higher grade steel is also great because it doesn’t really increase impacts and you can use less total material!
For concrete since it’s all about reducing cement, there are definitely some low hanging fruit without needing a product-specific EPD (though that’s still an important ask) - use portland limestone cement rather than portland cement, be more conscious about your cure times and don’t just use 28 days arbitrarily (foundations can often sit there for a while), specify maximum cement content lower than the NRMCA baseline in your region, use CarbonCure, SCM replacement (but keep in mind it’s not about % replacement but about total cement content). There’s also exciting new stuff like blue planet aggregate, but I don’t think that’s been used on a very wide basis beyond a small handful of projects in the Bay Area where they’re located. Not sure if that’s the answer you were looking for, but those are common strategies for reducing concrete impacts.