Hariom Gupta

Anmol Gulati, Hariom Gupta, Elias Lumer et al.

Long-horizon AI agents execute complex workflows spanning hundreds of sequential actions, yet a single wrong assumption early on can cascade into irreversible errors. When instructions are incomplete, the agent must decide not only whether to ask for clarification but when, and no prior work measures how clarification value changes over the course of execution. We introduce a forced-injection framework that provides ground-truth clarifications at controlled points in the agent's trajectory across four information dimensions (goal, input, constraint, context), three agent benchmarks, and four frontier models (three per benchmark; one on a single benchmark only; 84 task variants; 6,000+ runs). Counter to the common intuition that "earlier is always better," we find that the value of clarification depends sharply on what information is missing: goal clarification loses nearly all value after 10% of execution (pass@3 drops from 0.78 to baseline), while input clarification retains value through roughly 50%. Deferring any clarification type past mid-trajectory degrades performance below never asking at all. Cross-model Kendall tau correlations (0.78-0.87 among models sharing identical task coverage; 0.34-0.67 across the full 4-model panel) confirm these timing profiles are substantially task-intrinsic. A complementary study of 300 unscripted sessions reveals that no current frontier model asks within the empirically optimal window, with strategies ranging from over-asking (52% of sessions) to never asking at all. These empirical demand curves provide the quantitative foundation that existing theoretical frameworks require but have lacked, and establish concrete design targets for timing-aware clarification policies. Code and data will be publicly released.

Abhishek Kashyap, Megha Agarwal, Hariom Gupta

Copy-move forgery is one of the simple and effective operations to create forged images. Recently, techniques based on singular value decomposition (SVD) are widely used to detect copy-move forgery (CMF). Some approaches based on SVD are most acceptable to detect copy-move forgery but some copy-move forgery detection approaches can not produce satisfactory detection results. Sometimes these approaches may even produce error results. According to our observation, detection result produced using SVD depend highly on those parameters whose values are often determined with experiences. These values are only applicable to a few images, which limit their application. To solve this problem, a novel approach named as copy-move forgery detection using Cuckoo search algorithm (CMFD-CS) is proposed in this paper. CMFD-CS integrates the CS algorithm into SVD. It utilizes the CS algorithm to generate customized parameter values for images, which are used CMFD under block-based framework.

Abhishek Kashyap, Rajesh Singh Parmar, Megha Agrawal et al.

With the headway of the advanced image handling software and altering tools, a computerized picture can be effectively controlled. The identification of image manipulation is vital in light of the fact that an image can be utilized as legitimate confirmation, in crime scene investigation, and in numerous different fields. The image forgery detection techniques intend to confirm the credibility of computerized pictures with no prior information about the original image. There are numerous routes for altering a picture, for example, resampling, splicing, and copy-move. In this paper, we have examined different type of image forgery and their detection techniques; mainly we focused on pixel based image forgery detection techniques.