List your company

Open-domain long-horizon evaluation on FRAMES (test)

23.91AccuracyHiMPO
15.704417.834719.96522.0953Jun 15, 2026

Evaluation results

MethodLinks
HiMPOContext budget=8k ctx., Compression strategy=naive_recency2026.0623.91
HiMPOContext budget=4k ctx., Compression strategy=naive_recency2026.0623.67
HiMPOContext budget=full ctx.2026.0623.67
HiMPOContext budget=16k ctx., Compression strategy=mem_aware2026.0623.54
HiMPOContext budget=8k ctx., Compression strategy=mem_aware2026.0623.3
HiMPOContext budget=4k ctx., Compression strategy=mem_aware2026.0623.18
HiMPOContext budget=16k ctx., Compression strategy=naive_recency2026.0623.18
SUPOContext budget=8k ctx., Compression strategy=naive_recency2026.0621.6
SUPOContext budget=4k ctx., Compression strategy=naive_recency2026.0621.36
SUPOContext budget=16k ctx., Compression strategy=mem_aware2026.0621.36
MEMPOContext budget=8k ctx., Compression strategy=naive_recency2026.0621.24
SUPOContext budget=full ctx.2026.0621.24
MEMPOContext budget=4k ctx., Compression strategy=naive_recency2026.0621
SUPOContext budget=16k ctx., Compression strategy=naive_recency2026.0621
SFT onlyContext budget=8k ctx., Compression strategy=naive_recency2026.0620.87
MEMPOContext budget=16k ctx., Compression strategy=mem_aware2026.0620.87
MEMPOContext budget=full ctx.2026.0620.87
SUPOContext budget=8k ctx., Compression strategy=mem_aware2026.0620.87
SUPOContext budget=4k ctx., Compression strategy=mem_aware2026.0620.75
ReActContext budget=8k ctx., Compression strategy=naive_recency2026.0620.63
SFT onlyContext budget=4k ctx., Compression strategy=naive_recency2026.0620.63
MEMPOContext budget=16k ctx., Compression strategy=naive_recency2026.0620.63
SFT onlyContext budget=16k ctx., Compression strategy=mem_aware2026.0620.51
SFT onlyContext budget=full ctx.2026.0620.51
MEMPOContext budget=8k ctx., Compression strategy=mem_aware2026.0620.51
MEMPOContext budget=4k ctx., Compression strategy=mem_aware2026.0620.39
SFT onlyContext budget=16k ctx., Compression strategy=naive_recency2026.0620.27
SFT onlyContext budget=8k ctx., Compression strategy=mem_aware2026.0620.15
SFT onlyContext budget=4k ctx., Compression strategy=mem_aware2026.0620.02
ReActContext budget=16k ctx., Compression strategy=mem_aware2026.0618.57
ReActContext budget=4k ctx., Compression strategy=mem_aware2026.0617.84
ReActContext budget=4k ctx., Compression strategy=naive_recency2026.0617.23
ReActContext budget=full ctx.2026.0617.11
ReActContext budget=16k ctx., Compression strategy=naive_recency2026.0616.87
ReActContext budget=8k ctx., Compression strategy=mem_aware2026.0616.02