Papers with Code adds better research and benchmark discovery
’s team is reviving and has added several new features. Papers now show SOTA badges when their results place in the top three for a , making strong current results easier to spot. Paper ranking has also changed.
The new trending score combines GitHub star velocity with activity from linked models, datasets, and Spaces. This helps surface work such as IndexCache, which is described as a core technique behind the GLM-5.2 model. External evals are now supported, so readers can see results beyond the numbers reported in a paper itself.
Examples include FrontierSWE and PostTrainBench numbers for GLM-5.2, and results on the CritPt physics reasoning . More tasks, , and evals are being added over time, and is also available at paperswithco.de.
Key points
- added SOTA badges, a new trending score, and external evals.
- SOTA badges appear when a paper ranks in the top three for a .
- The new trending score uses GitHub star velocity plus activity from models, datasets, and Spaces.
- External evals make results easier to compare, including agent and reasoning .
- This helps narrow research candidates, but it does not prove real token or cost savings by itself.